Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track

Franck Dernoncourt, Daniel Preoţiuc-Pietro, Anastasia Shimorina (Editors)

Anthology ID:: 2024.emnlp-industry
Month:: November
Year:: 2024
Address:: Miami, Florida, US
Venue:: EMNLP
SIG:
Publisher:: Association for Computational Linguistics
URL:: https://aclanthology.org/2024.emnlp-industry
DOI:: 10.18653/v1/2024.emnlp-industry
Bib Export formats:: BibTeX MODS XML EndNote
PDF:: https://aclanthology.org/2024.emnlp-industry.pdf

pdf bib
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track
Franck Dernoncourt | Daniel Preoţiuc-Pietro | Anastasia Shimorina

pdf bib abs
Optimizing Entity Resolution in Voice Interfaces: An ASR-Aware Entity Reference Expansion Approach
Jiangning Chen | Ziyun Zhang | Qianli Hu

This paper tackles the challenges presented by Automatic Speech Recognition (ASR) errors in voice-based dialog systems, specifically, their adverse impact on Entity Resolution (ER) as a downstream task. Navigating the equilibrium between accuracy and online retrieval’s speed requirement proves challenging, particularly when limited data links the failed mentions to resolved entities. In this paper, we propose a entity reference expansion system, injecting pairs of failed mentions and resolved entity names into the knowledge graph, enhancing its awareness of unresolved mentions. To address data scarcity, we introduce a synthetic data generation approach aligned with noise patterns. This, combined with an ASR-Error-Aware Loss function, facilitates the training of a RoBERTa model, which filters failed mentions and extracts entity pairs for knowledge graph expansion. These designs confront obstacles related to ASR noise, data limitations, and online entity retrieval.

pdf bib abs
Two-tiered Encoder-based Hallucination Detection for Retrieval-Augmented Generation in the Wild
Ilana Zimmerman | Jadin Tredup | Ethan Selfridge | Joseph Bradley

Detecting hallucinations, where Large Language Models (LLMs) are not factually consistent with a Knowledge Base (KB), is a challenge for Retrieval-Augmented Generation (RAG) systems. Current solutions rely on public datasets to develop prompts or fine-tune a Natural Language Inference (NLI) model. However, these approaches are not focused on developing an enterprise RAG system; they do not consider latency, train or evaluate on production data, nor do they handle non-verifiable statements such as small talk or questions. To address this, we leverage the customer service conversation data of four large brands to evaluate existing solutions and propose a set of small encoder models trained on a new dataset. We find the proposed models to outperform existing methods and highlight the value of combining a small amount of in-domain data with public datasets.

pdf bib abs
The Program Testing Ability of Large Language Models for Code
Weimin Xiong | Yiwen Guo | Hao Chen

Recent development of large language models (LLMs) for code like CodeX and CodeT5+ shows promise in achieving code intelligence. Their ability of synthesizing program targeting a pre-defined algorithmic coding task has been intensively tested and verified on datasets including HumanEval and MBPP. Yet, evaluation of these LLMs from more perspectives (than just program synthesis) is also anticipated, considering their broad scope of applications. In this paper, we explore their ability of automatic test cases generation. We show intriguing observations and reveal how the quality of their generated test cases can be improved. Following recent work which uses generated test cases to enhance program synthesis, we further leverage our findings in improving the quality of the synthesized programs and show +11.77% and +4.22% higher code pass rates on HumanEval+ comparing with the GPT-3.5-turbo baseline and the recent state-of-the-art, respectively. Our code is publicly available at https://github.com/asdasxzxcq/TestCaseGen.

pdf bib abs
Salient Information Prompting to Steer Content in Prompt-based Abstractive Summarization
Lei Xu | Mohammed Asad Karim | Saket Dingliwal | Aparna Elangovan

Large language models (LLMs) can generate fluent summaries across domains using prompting techniques, reducing the effort required for summarization applications. However, crafting effective prompts that guide LLMs to generate summaries with the appropriate level of detail and writing style remains a challenge. In this paper, we explore the use of salient information extracted from the source document to enhance summarization prompts. We show that adding keyphrases in prompts can improve ROUGE F1 and recall, making the generated summaries more similar to the reference and more complete. The number of keyphrases can control the precision-recall trade-off. Furthermore, our analysis reveals that incorporating phrase-level salient information is superior to word- or sentence-level. However, the impact on summary faithfulness is not universally positive across LLMs. To enable this approach, we introduce Keyphrase Signal Extractor (SigExt), a lightweight model that can be finetuned to extract salient keyphrases. By using SigExt, we achieve consistent ROUGE improvements across datasets and LLMs without any LLM customization. Our findings provide insights into leveraging salient information in building prompt-based summarization systems.

pdf bib abs
Predicting Entity Salience in Extremely Short Documents
Benjamin Bullough | Harrison Lundberg | Chen Hu | Weihang Xiao

A frequent challenge in applications that use entities extracted from text documents is selecting the most salient entities when only a small number can be used by the application (e.g., displayed to a user). Solving this challenge is particularly difficult in the setting of extremely short documents, such as the response from a digital assistant, where traditional signals of salience such as position and frequency are less likely to be useful. In this paper, we propose a lightweight and data-efficient approach for entity salience detection on short text documents. Our experiments show that our approach achieves competitive performance with respect to complex state-of-the-art models, such as GPT-4, at a significant advantage in latency and cost. In limited data settings, we show that a semi-supervised fine-tuning process can improve performance further. Furthermore, we introduce a novel human-labeled dataset for evaluating entity salience on short question-answer pair documents.

pdf bib abs
Don’t Shoot The Breeze: Topic Continuity Model Using Nonlinear Naive Bayes With Attention
Shu-Ting Pi | Pradeep Bagavan | Yejia Li | Disha Disha | Qun Liu

Utilizing Large Language Models (LLM) as chatbots in diverse business scenarios often presents the challenge of maintaining topic continuity. Abrupt shifts in topics can lead to poor user experiences and inefficient utilization of computational resources. In this paper, we present a topic continuity model aimed at assessing whether a response aligns with the initial conversation topic. Our model is built upon the expansion of the corresponding natural language understanding (NLU) model into quantifiable terms using a Naive Bayes approach. Subsequently, we have introduced an attention mechanism and logarithmic nonlinearity to enhance its capability to capture topic continuity. This approach allows us to convert the NLU model into an interpretable analytical formula. In contrast to many NLU models constrained by token limits, our proposed model can seamlessly handle conversations of any length with linear time complexity. Furthermore, the attention mechanism significantly improves the model’s ability to identify topic continuity in complex conversations. According to our experiments, our model consistently outperforms traditional methods, particularly in handling lengthy and intricate conversations. This unique capability offers us an opportunity to ensure the responsible and interpretable use of LLMs.

pdf bib abs
Retrieval Augmented Spelling Correction for E-Commerce Applications
Xuan Guo | Rohit Patki | Dante Everaert | Christopher Potts

The rapid introduction of new brand names into everyday language poses a unique challenge for e-commerce spelling correction services, which must distinguish genuine misspellings from novel brand names that use unconventional spelling. We seek to address this challenge via Retrieval Augmented Generation (RAG). On this approach, product names are retrieved from a catalog and incorporated into the context used by a large language model (LLM) that has been fine-tuned to do contextual spelling correction. Through quantitative evaluation and qualitative error analyses, we find improvements in spelling correction utilizing the RAG framework beyond a stand-alone LLM. We also demonstrate the value of additional finetuning of the LLM to incorporate retrieved context.

Scaling laws in language modeling traditionally quantify training loss as a function of dataset size and model parameters, providing compute-optimal estimates but often neglecting the impact of data quality on model generalization.In this paper, we extend the conventional understanding of scaling law by offering a microscopic view of data quality within the original formulation – effective training tokens – which we posit to be a critical determinant of performance for parameter-constrained language models.Specifically, we formulate the proposed term of effective training tokens to be a combination of two readily-computed indicators of text:(i) text diversity and (ii) syntheticity as measured by a teacher model.We pretrained over 200 models of 25M to 1.5B parameters on a diverse set of sampled, synthetic data, and estimated the constants that relate text quality, model size, training tokens, and eight reasoning task accuracy scores.We demonstrated the estimated constants yield +0.83 Pearson correlation with true accuracies, and analyze it in scenarios involving widely-used data techniques such as data sampling and synthesis which aim to improve data quality.

Large language models (LLMs) trained on general domain corpora showed remarkable results on natural language processing (NLP) tasks. However, previous research demonstrated LLMs trained using domain-focused corpora perform better on specialized tasks. Inspired by this insight, we developed INDUS, a comprehensive suite of LLMs tailored for the closely-related domains of Earth science, biology, physics, heliophysics, planetary sciences and astrophysics, and trained using curated scientific corpora drawn from diverse data sources. The suite of models include: (1) an encoder model trained using domain-specific vocabulary and corpora to address NLP tasks, (2) a contrastive-learning based text embedding model trained using a diverse set of datasets to address information retrieval tasks and (3) smaller versions of these models created using knowledge distillation for applications which have latency or resource constraints. We also created three new scientific benchmark datasets, Climate-Change NER (entity-recognition), NASA-QA (extractive QA) and NASA-IR (IR) to accelerate research in these multi-disciplinary fields. We show that our models outperform both general-purpose (RoBERTa) and domain- specific (SciBERT) encoders on these new tasks as well as existing tasks in the domains of interest. Furthermore, we demonstrate the use of these models in two industrial settings- as a retrieval model for large-scale vector search applications and in automatic content tagging systems.

pdf bib abs
DL-QAT: Weight-Decomposed Low-Rank Quantization-Aware Training for Large Language Models
Wenjing Ke | Zhe Li | Dong Li | Lu Tian | Emad Barsoum

Improving the efficiency of inference in Large Language Models (LLMs) is a critical area of research. Post-training Quantization (PTQ) is a popular technique, but it often faces challenges at low-bit levels, particularly in downstream tasks. Quantization-aware Training (QAT) can alleviate this problem, but it requires significantly more computational resources. To tackle this, we introduced Weight-Decomposed Low-Rank Quantization-Aware Training (DL-QAT), which merges the advantages of QAT while training only less than 1% of the total parameters. Specifically, we introduce a group-specific quantization magnitude to adjust the overall scale of each quantization group. Within each quantization group, we use LoRA matrices to update the weight size and direction in the quantization space. We validated the effectiveness of our method on the LLaMA and LLaMA2 model families. The results show significant improvements over our baseline method across different quantization granularities. For instance, for LLaMA-7B, our approach outperforms the previous state-of-the-art method by 4.2% in MMLU on 3-bit LLaMA-7B. Additionally, our quantization results on pre-trained models also surpass previous QAT methods, demonstrating the superior performance and efficiency of our approach.

Large language models (LLMs) enhanced with retrieval augmentation has shown great performance in many applications. However, the computational demands for these models pose a challenge when applying them to real-time tasks, such as composition assistance. To address this, we propose Hybrid Retrieval-Augmented Composition Assistance (Hybrid-RACA), a novel system for real-time text prediction that efficiently combines a cloud-based LLM with a smaller client-side model through retrieval augmented memory. This integration enables the client model to generate better responses, benefiting from the LLM’s capabilities and cloud-based data. Meanwhile, via a novel asynchronous memory update mechanism, the client model can deliver real-time completions to user inputs without the need to wait for responses from the cloud. Our experiments on five datasets demonstrate that Hybrid-RACA offers strong performance while maintaining low latency.

Recent advancements in large language models (LLMs) are propelling us toward artificial general intelligence with their remarkable emergent abilities and reasoning capabilities. However, the substantial computational and memory requirements limit the widespread adoption. Quantization, a key compression technique, can effectively mitigate these demands by compressing and accelerating LLMs, albeit with potential risks to accuracy. Numerous studies have aimed to minimize the accuracy loss associated with quantization. However, their quantization configurations vary from each other and cannot be fairly compared. In this paper, we present LLMC, a plug-and-play compression toolkit, to fairly and systematically explore the impact of quantization. LLMC integrates dozens of algorithms, models, and hardware, offering high extensibility from integer to floating-point quantization, from LLM to vision-language (VLM) model, from fixed-bit to mixed precision, and from quantization to sparsification. Powered by this versatile toolkit, our benchmark covers three key aspects: calibration data, algorithms (three strategies), and data formats, providing novel insights and detailed analyses for further research and practical guidance for users. Our toolkit is available at https://github.com/ModelTC/llmc.

Large Language Models (LLMs) have issues with document question answering (QA) in situations where the document is unable to fit in the small context length of an LLM. To overcome this issue, most existing works focus on retrieving the relevant context from the document, representing them as plain text. However, documents such as PDFs, web pages, and presentations are naturally structured with different pages, tables, sections, and so on. Representing such structured documents as plain text is incongruous with the user’s mental model of these documents with rich structure. When a system has to query the document for context, this incongruity is brought to the fore, and seemingly trivial questions can trip up the QA system. To bridge this fundamental gap in handling structured documents, we propose an approach called PDFTriage that enables models to retrieve the context based on either structure or content. Our experiments demonstrate the effectiveness of the proposed PDFTriage-augmented models across several classes of questions where existing retrieval-augmented LLMs fail. To facilitate further research on this fundamental problem, we release our benchmark dataset consisting of 900+ human-generated questions over 80 structured documents from 10 different categories of question types for document QA. Our code and datasets will be released soon on Github.

pdf bib abs
Fairness-Aware Online Positive-Unlabeled Learning
Hoin Jung | Xiaoqian Wang

Machine learning applications for text classification are increasingly used in domains such as toxicity and misinformation detection in online settings. However, obtaining precisely labeled data for training remains challenging, particularly because not all problematic instances are reported. Positive-Unlabeled (PU) learning, which uses only labeled positive and unlabeled samples, offers a solution for these scenarios. A significant concern in PU learning, especially in online settings, is fairness: specific groups may be disproportionately classified as problematic. Despite its importance, this issue has not been explicitly addressed in research. This paper aims to bridge this gap by investigating the fairness of PU learning in both offline and online settings. We propose a novel approach to achieve more equitable results by extending PU learning methods to online learning for both linear and non-linear classifiers and analyzing the impact of the online setting on fairness. Our approach incorporates a convex fairness constraint during training, applicable to both offline and online PU learning. Our solution is theoretically robust, and experimental results demonstrate its efficacy in improving fairness in PU learning in text classification.

This study presents a novel learning approach designed to enhance both mathematical reasoning and problem-solving abilities of Large Language Models (LLMs). We focus on integrating the Chain-of-Thought (CoT) and the Program-of-Thought (PoT) learning, hypothesizing that prioritizing the learning of mathematical reasoning ability is helpful for the amplification of problem-solving ability. Thus, the initial learning with CoT is essential for solving challenging mathematical problems. To this end, we propose a sequential learning approach, named SAAS (Solving Ability Amplification Strategy), which strategically transitions from CoT learning to PoT learning. Our empirical study, involving an extensive performance comparison using several benchmarks, demonstrates that our SAAS achieves state-of-the-art (SOTA) performance. The results underscore the effectiveness of our sequential learning approach, marking a significant advancement in the field of mathematical reasoning in LLMs.

pdf bib abs
Debiasing Text Safety Classifiers through a Fairness-Aware Ensemble
Olivia Sturman | Aparna R Joshi | Bhaktipriya Radharapu | Piyush Kumar | Renee Shelby

Increasing use of large language models (LLMs) demand performant guardrails to ensure the safety of inputs and outputs of LLMs. When these safeguards are trained on imbalanced data, they can learn the societal biases. We present a light-weight, post-processing method for mitigating counterfactual fairness in closed-source text safety classifiers. Our approach involves building an ensemble that not only outperforms the input classifiers and policy-aligns them, but also acts as a debiasing regularizer. We introduce two threshold-agnostic metrics to assess the counterfactual fairness of a model, and demonstrate how combining these metrics with Fair Data Reweighting (FDW) helps mitigate biases. We create an expanded Open AI dataset, and a new templated LLM-generated dataset based on user-prompts, both of which are counterfactually balanced across identity groups and cover four key areas of safety; we will work towards publicly releasing these datasets. Our results show that our approach improves counterfactual fairness with minimal impact on model performance.

This paper addresses the challenge of improving user experience on e-commerce platforms by enhancing product ranking relevant to user’s search queries. Ambiguity and complexity of user queries often lead to a mismatch between user’s intent and retrieved product titles or documents. Recent approaches have proposed the use of Transformer-based models which need millions of annotated query-title pairs during the pre-training stage, and this data often does not take user intent into account. To tackle this, we curate samples from existing datasets at eBay, manually annotated with buyer-centric relevance scores, and centrality scores which reflect how well the product title matches the user’s intent. We introduce a User-intent Centrality Optimization (UCO) approach for existing models, which optimizes for the user intent in semantic product search. To that end, we propose a dual-loss based optimization to handle hard negatives, i.e., product titles that are semantically relevant but do not reflect the user’s intent. Our contributions include curating challenging evaluation sets and implementing UCO, resulting in significant improvements in product ranking efficiency, observed for different evaluation metrics. Our work aims to ensure that the most buyer-centric titles for a query are ranked higher, thereby, enhancing the user experience on e-commerce platforms.

Evaluating natural language generation (NLG) systems automatically poses significant challenges.Recent studies have employed large language models (LLMs) as reference-free metrics for NLG evaluation, enhancing adaptability to new tasks tasks. However, these methods still show lower correspondence with human judgments compared to specialized neural evaluators.In this paper, we introduce “Fusion-Eval”, an innovative approach that leverages LLMs to integrate insights from various assistant evaluators. The LLM is given the example to evaluate along with scores from the assistant evaluators. Each of these evaluators specializes in assessing distinct aspects of responses.Fusion-Eval achieves a 0.962 system-level Kendall-Tau correlation with humans on SummEval and a 0.744 turn-level Spearman correlation on TopicalChat, which is significantly higher than baseline methods. These results highlight Fusion-Eval’s significant potential in the realm of natural language system evaluation.

pdf bib abs
Investigating the Personality Consistency in Quantized Role-Playing Dialogue Agents
Yixiao Wang | Homa Fashandi | Kevin Ferreira

This study explores the consistency of personality traits in quantized large language models (LLMs) for edge device role-playing scenarios. Using the Big Five personality traits model, we evaluate how stable assigned personalities are for Quantized Role-Playing Dialog Agents (QRPDA) during multi-turn interactions. We evaluate multiple LLMs with various quantization levels, combining binary indexing of personality traits, explicit self-assessments, and linguistic analysis of narratives. To address personality inconsistency, we propose a non-parametric method called Think2. Our multi-faceted evaluation framework demonstrates Think2’s effectiveness in maintaining consistent personality traits for QRPDA. Moreover, we offer insights to help select the optimal model for QRPDA, improving its stability and reliability in real-world applications.

pdf bib abs
Robust ASR Error Correction with Conservative Data Filtering
Takuma Udagawa | Masayuki Suzuki | Masayasu Muraoka | Gakuto Kurata

Error correction (EC) based on large language models is an emerging technology to enhance the performance of automatic speech recognition (ASR) systems.Generally, training data for EC are collected by automatically pairing a large set of ASR hypotheses (as sources) and their gold references (as targets).However, the quality of such pairs is not guaranteed, and we observed various types of noise which can make the EC models brittle, e.g. inducing overcorrection in out-of-domain (OOD) settings.In this work, we propose two fundamental criteria that EC training data should satisfy: namely, EC targets should (1) improve linguistic acceptability over sources and (2) be inferable from the available context (e.g. source phonemes).Through these criteria, we identify low-quality EC pairs and train the models not to make any correction in such cases, the process we refer to as conservative data filtering.In our experiments, we focus on Japanese ASR using a strong Conformer-CTC as the baseline and finetune Japanese LLMs for EC.Through our evaluation on a suite of 21 internal benchmarks, we demonstrate that our approach can significantly reduce overcorrection and improve both the accuracy and quality of ASR results in the challenging OOD settings.

Language models for natural language processing have been grafted onto programming language modeling for advancing code intelligence. Although it can be represented in the text format, code is syntactically more rigorous, as it is designed to be properly compiled or interpreted to perform a set of behaviors given any inputs. In this case, existing works benefit from syntactic representations to learn from code less ambiguously in forms of abstract syntax tree, control-flow graph, etc. However, programs with the same purpose can be implemented in various ways showing different syntactic representations, while the ones with similar implementations can have distinct behaviors. Though trivially demonstrated during executions, such semantics about functionality are challenging to be learned directly from code, especially in an unsupervised manner. Hence, in this paper, we propose FuzzPretrain to explore the dynamic information of programs revealed by their test cases and embed it into the feature representations of code as complements. The test cases are obtained with the assistance of a customized fuzzer and are only required during pre-training. FuzzPretrain yielded more than 6%/19% mAP improvements on code search over its masked language modeling counterparts trained with only source code and source code coupled with abstract syntax trees (ASTs), respectively. Our experiments show the benefits of learning discriminative code representations from FuzzPretrain.

Large language models (LLMs) have surged in popularity and are extensively used in commercial applications, where the efficiency of model serving is crucial for the user experience. Most current research focuses on optimizing individual sub-procedures, e.g. local inference and communication, however, there is no comprehensive framework that provides a holistic system view for optimizing LLM serving in an end-to-end manner. In this work, we conduct a detailed analysis to identify major bottlenecks that impact end-to-end latency in LLM serving systems. Our analysis reveals that a comprehensive LLM serving endpoint must address a series of efficiency bottlenecks that extend beyond LLM inference. We then propose ScaleLLM, an optimized system for resource-efficient LLM serving. Our extensive experiments reveal that reveal that with 64 concurrent requests on Mixtral 8x7B, ScaleLLM achieves a 4.3× speed up over vLLM and outperforms state-of-the-arts with 1.5× higher throughput.

This paper introduces a novel framework that combines graph-driven context retrieval in conjunction to knowledge graphs based enhancement, honing the proficiency of LLMs, especially in domain specific community question answering platforms like AskUbuntu, Unix, and ServerFault. We conduct experiments on various LLMs with different parameter sizes to evaluate their ability to ground knowledge and determine factual accuracy in answers to open-ended questions. Our methodology GraphContextGen consistently outperforms dominant text-based retrieval systems, demonstrating its robustness and adaptability to a larger number of use cases. This advancement highlights the importance of pairing context rich data retrieval with LLMs, offering a renewed approach to knowledge sourcing and generation in AI systems. We also show that, due to rich contextual data retrieval, the crucial entities, along with the generated answer, remain factually coherent with the gold answer. We shall release the source code and datasets upon acceptance.

The electric vehicle (EV) battery supply chain’s vulnerability to disruptions necessitates advanced predictive analytics. We present SHIELD (Schema-based Hierarchical Induction for EV supply chain Disruption), a system integrating Large Language Models (LLMs) with domain expertise for EV battery supply chain risk assessment. SHIELD combines: (1) LLM-driven schema learning to construct a comprehensive knowledge library, (2) a disruption analysis system utilizing fine-tuned language models for event extraction, multi-dimensional similarity matching for schema matching, and Graph Convolutional Networks (GCNs) with logical constraints for prediction, and (3) an interactive interface for visualizing results and incorporating expert feedback to enhance decision-making. Evaluated on 12,070 paragraphs from 365 sources (2022-2023), SHIELD outperforms baseline GCNs and LLM+prompt methods (e.g. GPT-4o) in disruption prediction. These results demonstrate SHIELD’s effectiveness in combining LLM capabilities with domain expertise for enhanced supply chain risk assessment.

Evaluating the quality and consistency of text generated by Large Language Models (LLMs) poses a significant, yet unresolved challenge for industry research. We propose , an automated framework for evaluating and improving the consistency of LLM-generated texts using a divide-conquer-reasoning approach. Unlike existing LLM-based evaluators operating at the paragraph level, our method employs a divide-and-conquer evaluator () that breaks down the paragraph-to-paragraph comparison into sentence-to-paragraph comparisons. To facilitate this approach, we also introduce an automatic metric converter () that translates the output from into an interpretable numeric score. Beyond the consistency evaluation, we further present a reason-assisted improver () that mitigates inconsistencies by leveraging the analytical reasons identified by . Through comprehensive and systematic empirical analysis, we show that our approach outperforms state-of-the-art methods by a large margin (e.g., +16.8% and +32.5% on the SummEval dataset) in consistency evaluation across multiple benchmarks. Our approach also substantially reduces nearly 90% output inconsistencies in one iteration, showing promise for effective hallucination mitigation in real-world industrial applications.

pdf bib abs
Greenback Bears and Fiscal Hawks: Finance is a Jungle and Text Embeddings Must Adapt
Peter Anderson | Mano Vikash Janardhanan | Jason He | Wei Cheng | Charlie Flanagan

Financial documents are filled with specialized terminology, arcane jargon, and curious acronyms that pose challenges for general-purpose text embeddings. Yet, few text embeddings specialized for finance have been reported in the literature, perhaps in part due to a lack of public datasets and benchmarks. We present BAM embeddings, a set of text embeddings finetuned on a carefully constructed dataset of 14.3M query-passage pairs including both public and proprietary financial documents. Demonstrating the benefits of domain-specific training, BAM embeddings achieve Recall@1 of 62.8% on a held-out test set, vs. only 39.2% for the best general-purpose text embedding from OpenAI. Further, BAM embeddings increase question answering accuracy by 8% on FinanceBench and show increased sensitivity to the finance-specific elements that are found in detailed, forward-looking and company and date-specific queries. To support further research we describe our approach in detail, quantify the importance of hard negative mining and dataset scale, and publicly release our embeddings.

Large Language Models (LLMs) have demonstrated proficiency in addressing tasks that necessitate a combination of task planning and the usage of external tools, such as weather and calculator APIs. However, real-world industrial systems present prevalent challenges in task planning and tool usage: numerous APIs in the real system make it intricate to invoke the appropriate one, while the inherent limitations of LLMs pose challenges in orchestrating an accurate sub-task sequence and API-calling order. This paper introduces a comprehensive framework aimed at enhancing the Task Planning and Tool Usage (TPTU) abilities of LLM-based agents in industry. Our framework comprises three key components designed to address these challenges: (1) the API Retriever selects the most pertinent APIs among the extensive API set; (2) the Demo Selector retrieves task-level demonstrations, which is further used for in-context learning to aid LLMs in accurately decomposing subtasks and effectively invoking hard-to-distinguish APIs; (3) LLM Finetuner tunes a base LLM to enhance its capability for task planning and API calling. We validate our methods using a real-world industry system and an open-sourced academic dataset, demonstrating the efficacy of each individual component as well as the integrated framework. The code is available at here.

pdf bib abs
Detecting Ambiguous Utterances in an Intelligent Assistant
Satoshi Akasaki | Manabu Sassano

In intelligent assistants that perform both chatting and tasks through dialogue, like Siri and Alexa, users often make ambiguous utterances such as “I’m hungry” or “I have a headache,” which can be interpreted as either chat or task intents. Naively determining these intents can lead to mismatched responses, spoiling the user experience. Therefore, it is desirable to determine the ambiguity of user utterances. We created a dataset from an actual intelligent assistant via crowdsourcing and analyzed tendencies of ambiguous utterances. Using this labeled data of chat, task, and ambiguous intents, we developed a supervised intent classification model. To detect ambiguous utterances robustly, we propose feeding sentence embeddings developed from microblogs and search logs with a self-attention mechanism. Experiments showed that our model outperformed two baselines, including a strong LLM-based one. We will release the dataset.

pdf bib abs
GeoIndia: A Seq2Seq Geocoding Approach for Indian Addresses
Bhavuk Singhal | Anshu Aditya | Lokesh Todwal | Shubham Jain | Debashis Mukherjee

Geocoding, the conversion of unstructured geographic text into structured spatial data, is essential for logistics, urban planning, and location-based services. Indian addresses with their diverse languages, scripts, and formats present significant challenges that existing geocoding methods often fail to address, particularly at fine-grained resolutions. In this paper, we propose GeoIndia, a novel geocoding system designed specifically for Indian addresses using hierarchical H3-cell prediction within a Seq2Seq framework. Our methodology includes a comprehensive analysis of Indian addressing systems, leading to the development of a data correction strategy that enhances prediction accuracy. We investigate two model architectures, Flan-T5-base (T5) and Llama-3-8b (QLF-Llama-3), due to their strong sequence generation capabilities. We trained around 29 models with one dedicated to each state, and results show that our approach provides superior accuracy and reliability across multiple Indian states, outperforming the well-renowned geocoding platform Google Maps. In multiple states, we achieved more than an 50% reduction in mean distance error and more than a 85% reduction in 99th percentile distance error compared to Google Maps. This advancement can help in optimizing logistics in the e-commerce sector, reducing delivery failures and improving customer satisfaction.

pdf bib abs
Moleco: Molecular Contrastive Learning with Chemical Language Models for Molecular Property Prediction
Jun-Hyung Park | Hyuntae Park | Yeachan Kim | Woosang Lim | SangKeun Lee

Pre-trained chemical language models (CLMs) excel in the field of molecular property prediction, utilizing string-based molecular descriptors such as SMILES for learning universal representations. However, such string-based descriptors implicitly contain limited structural information, which is closely associated with molecular property prediction. In this work, we introduce Moleco, a novel contrastive learning framework to enhance the understanding of molecular structures within CLMs. Based on the similarity of fingerprint vectors among different molecules, we train CLMs to distinguish structurally similar and dissimilar molecules in a contrastive manner. Experimental results demonstrate that Moleco significantly improves the molecular property prediction performance of CLMs, outperforming state-of-the-art models. Moreover, our in-depth analysis with diverse Moleco variants verifies that fingerprint vectors are highly effective features in improving CLMs’ understanding of the structural information of molecules.

Materials science is an interdisciplinary field focused on studying and discovering materials around us. However, due to the vast space of materials, datasets in this field are typically scarce and have limited coverage. This inherent limitation makes current adaptation methods less effective when adapting pre-trained language models (PLMs) to materials science, as these methods rely heavily on the frequency information from limited downstream datasets. In this paper, we propose Semantic Knowledge Transfer (SEED), a novel vocabulary expansion method to adapt the pre-trained language models for materials science. The core strategy of SEED is to transfer the materials knowledge of lightweight embeddings into the PLMs. To this end, we introduce knowledge bridge networks, which learn to transfer the latent knowledge of the materials embeddings into ones compatible with PLMs. By expanding the embedding layer of PLMs with these transformed embeddings, PLMs can comprehensively understand the complex terminology associated with materials science. We conduct extensive experiments across a broad range of materials-related benchmarks. Comprehensive evaluation results convincingly demonstrate that SEED mitigates the mentioned limitations of previous adaptation methods, showcasing the efficacy of transferring embedding knowledge into PLMs.

Credit risk monitoring is an essential process for financial institutions to evaluate the creditworthiness of borrowing entities and minimize potential losses. Traditionally, this involves the periodic assessment of news regarding client companies to identify events which can impact their financial standing. This process can prove arduous and delay a timely response to credit impacting events. The News Risk Alerting System (NRAS) proactively identifies credit-relevant news related to clients and alerts the relevant Credit Officer (CO). This production system has been deployed for nearly three years and has alerted COs to over 2700 credit-relevant events with an estimated precision of 77%.

pdf bib abs
FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model
Yichen Lu | Jiaqi Song | Chao-Han Huck Yang | Shinji Watanabe

In this study, we aim to explore Multitask Speech Language Model (SpeechLM) efficient inference via token reduction. Unlike other modalities such as vision or text, speech has unique temporal dependencies, making previous efficient inference works on other modalities not directly applicable. Furthermore, methods for efficient SpeechLM inference on long sequence and sparse signals remain largely unexplored. In this work, we propose FastAdaSP, a weighted token merging framework specifically designed for various speech-related tasks to improve the trade-off between efficiency and performance. Experimental results on WavLLM and Qwen-Audio show that our method achieves the state-of-the-art (SOTA) efficiency-performance trade-off compared with other baseline methods. Specifically, FastAdaSP achieved 7x memory efficiency and 1.83x decoding throughput without any degradation on tasks like Emotion Recognition (ER) and Spoken Question Answering (SQA).

With the rapid growth of Large Language Models (LLMs) across various domains, numerous new LLMs have emerged, each possessing domain-specific expertise. This proliferation has highlighted the need for quick, high-quality, and cost-effective LLM query response methods. Yet, no single LLM exists to efficiently balance this trilemma. Some models are powerful but extremely costly, while others are fast and inexpensive but qualitatively inferior. To address this challenge, we present TO-Router, a non-monolithic LLM querying system that seamlessly integrates various LLM experts into a single query interface and dynamically routes incoming queries to the most high-performant expert based on query’s requirements. Through extensive experiments, we demonstrate that when compared to standalone expert models, TO-Router improves query efficiency by up to 40%, and leads to significant cost reductions of up to 30%, while maintaining or enhancing model performance by up to 10%.

pdf bib abs
Prompt-Tuned Muti-Task Taxonomic Transformer (PTMTTaxoFormer)
Rajashekar Vasantha | Nhan Nguyen | Yue Zhang

Hierarchical Text Classification (HTC) is a subclass of multi-label classification, it is challenging because the hierarchy typically has a large number of diverse topics. Existing methods for HTC fall within two categories, local methods (a classifier for each level, node, or parent) or global methods (a single classifier for everything). Local methods are computationally expensive, whereas global methods often require complex explicit injection of the hierarchy, verbalizers, and/or prompt engineering. In this work, we propose Prompt Tuned Multi Task Taxonomic Transformer, a single classifier that uses a multi-task objective to predict one or more topics. The approach is capable of understanding the hierarchy during training without explicit injection, complex heads, verbalizers, or prompt engineering. PTMTTaxoFormer is a novel model architecture and training paradigm using differentiable prompts and labels that are learnt through backpropagation. PTMTTaxoFormer achieves state of the art results on several HTC benchmarks that span a range of topics consistently. Compared to most other HTC models, it has a simpler yet effective architecture, making it more production-friendly in terms of latency requirements (a factor of 2-5 lower latency). It is also robust and label-efficient, outperforming other models with 15%-50% less training data.

The rapid growth of open-source language models provides the opportunity to merge model checkpoints, combining their parameters to improve performance and versatility. Advances in transfer learning have led to numerous task-specific models, which model merging can integrate into powerful multitask models without additional training. MergeKit is an open-source library designed to support this process with an efficient and extensible framework suitable for any hardware. It has facilitated the merging of thousands of models, contributing to some of the world’s most powerful open-source model checkpoints. The library is accessible at: https://github.com/arcee-ai/mergekit.

Large Language Models (LLMs) have made significant progress, becoming more autonomous and capable of handling real-world tasks through their access to tools, various planning strategies, and memory, referred to as LLM agents. One emerging area of focus is customizing these models to cater to individual user preferences, thereby shaping them into personal LLM agents. This work investigates how the user model, which encapsulates user-related information, preferences, and personal concepts, influences an LLM agent’s planning and reasoning capabilities. We introduce a personalized version of TravelPlanner, called TravelPlanner+, and establish baselines for personal LLM agents. Our evaluation strategy contains an LLM-as-a-Judge component, which provides further in-depth insights into the decision-making process of a personal LLM agent by comparing generic and personal plans. Our findings reveal that while generic plans perform robustly, personal plans show marked improvement in relevance and suitability, with preference rates up to 74.4% on validation and 87.3% on the test set. These results highlight the potential of personal LLM agents to significantly enhance user satisfaction.

Full-parameter fine-tuning is computationally prohibitive for large language models (LLMs), making parameter-efficient fine-tuning (PEFT) methods like low-rank adaptation (LoRA) increasingly popular. However, LoRA and its existing variants introduce significant latency in multi-tenant settings, hindering their applications in the industry. To address this issue, we propose the Fantastic LoRA (FanLoRA) framework, which consists of four steps: (a) adding LoRA modules to all the Transformer linear weights and fine-tuning on a large-scale instruction tuning dataset. (b) The importance of each module is then assessed using a novel importance scoring method. (c) only the most critical modules per layer are retained, resulting in the FanLoRA setting. (d) The FanLoRA setting is applied to fine-tune various downstream tasks. Our extensive experiments demonstrate that: (a) FanLoRA outperforms existing PEFT baselines across a wide collection of tasks with comparable tunable parameters. (b) FanLoRA significantly reduces the inference latency of LoRA, making it valuable for further broadening the applications of LLMs in the industry.

pdf bib abs
ReportGPT: Human-in-the-loop Verifiable Table-to-Text Generation
Lucas Cecchi | Petr Babkin

Recent developments in the quality and accessibility of large language models have precipitated a surge in user-facing tools for content generation. Motivated by a necessity for human quality control of these systems, we introduce ReportGPT: a pipeline framework for verifiable human-in-the-loop table-to-text generation. ReportGPT is based on a domain specific language, which acts as a proof mechanism for generating verifiable commentary. This allows users to quickly check the relevancy and factuality of model outputs. User selections then become few-shot examples for improving the performance of the pipeline. We configure 3 approaches to our pipeline, and find that usage of language models in ReportGPT’s components trade off precision for more insightful downstream commentary. Furthermore, ReportGPT learns from human feedback in real-time, needing only a few samples to improve performance.

Data deduplication is a critical task in data management and mining, focused on consolidating duplicate records that refer to the same entity. Personally Identifiable Information (PII) is a critical class of data for deduplication across various industries. Consumer data, stored and generated through various engagement channels, is crucial for marketers, agencies, and publishers. However, a major challenge to PII data deduplication is the lack of open-source benchmark datasets due to stringent privacy concerns, which hinders the research, development, and evaluation of robust solutions.This paper addresses this critical lack of PII deduplication benchmarks by introducing the first open-source, high-quality dataset for this task. We provide two datasets: one with 1,000,000 unlabeled synthetic PII profiles and a subset of 10,000 pairs curated and labeled by trained annotators as matches or non-matches. Our datasets contain synthetic profiles built from publicly available sources that do not represent any real individuals, thus ensuring privacy and ethical compliance. We provide several challenging data variations to evaluate the effectiveness of various deduplication techniques, including traditional supervised methods, deep-learning approaches, and large language models (LLMs). Our work aims to set a new standard for PII deduplication, paving the way for more accurate and secure solutions. We share our data publicly at this link - https://zenodo.org/records/13932202.

pdf bib abs
MERLIN: Multimodal Embedding Refinement via LLM-based Iterative Navigation for Text-Video Retrieval-Rerank Pipeline
Donghoon Han | Eunhwan Park | Gisang Lee | Adam Lee | Nojun Kwak

The rapid expansion of multimedia content has made accurately retrieving relevant videos from large collections increasingly challenging. Recent advancements in text-video retrieval have focused on cross-modal interactions, large-scale foundation model training, and probabilistic modeling, yet often neglect the crucial user perspective, leading to discrepancies between user queries and the content retrieved. To address this, we introduce MERLIN (Multimodal Embedding Refinement via LLM-based Iterative Navigation), a novel, training-free pipeline that leverages Large Language Models (LLMs) for iterative feedback learning. MERLIN refines query embeddings from a user perspective, enhancing alignment between queries and video content through a dynamic question answering process. Experimental results on datasets like MSR-VTT, MSVD, and ActivityNet demonstrate that MERLIN substantially improves Recall@1, outperforming existing systems and confirming the benefits of integrating LLMs into multimodal retrieval systems for more responsive and context-aware multimedia retrieval.

pdf bib abs
Identifying High Consideration E-Commerce Search Queries
Zhiyu Chen | Jason Ingyu Choi | Besnik Fetahu | Shervin Malmasi

In e-commerce, high consideration search missions typically require careful and elaborate decision making, and involve a substantial research investment from customers. We consider the task of identifying High Consideration (HC) queries. Identifying such queries enables e-commerce sites to better serve user needs using targeted experiences such as curated QA widgets that help users reach purchase decisions. We explore the task by proposing an Engagement-based Query Ranking (EQR) approach, focusing on query ranking to indicate potential engagement levels with query-related shopping knowledge content during product search. Unlike previous studies on predicting trends, EQR prioritizes query-level features related to customer behavior, finance, and catalog information rather than popularity signals. We introduce an accurate and scalable method for EQR and present experimental results demonstrating its effectiveness. Offline experiments show strong ranking performance. Human evaluation shows a precision of 96% for HC queries identified by our model. The model was commercially deployed, and shown to outperform human-selected queries in terms of downstream customer impact, as measured through engagement.

Large language models (LLMs) have achieved significant leadership in many NLP tasks, but aligning structured output with generative models in information extraction (IE) tasks remains a challenge. Prompt Engineering (PE) is renowned for improving IE performance through prompt modifications. However, the realm of the sample design for downstream fine-tuning, crucial for task-specific LLM adaptation, is largely unexplored. This paper introduces **Sample Design Engineering** (SDE), a methodical approach to enhancing LLMs’ post-tuning performance on IE tasks by refining input, output, and reasoning designs. Through extensive ID and OOD experiments across six LLMs, we first assess the impact of various design options on IE performance, revealing several intriguing patterns. Based on these insights, we then propose an integrated SDE strategy and validate its consistent superiority over heuristic sample designs on three complex IE tasks with four additional LLMs, demonstrating the generality of our method. Additionally, analyses of LLMs’ inherent prompt/output perplexity, zero-shot, and ICL abilities illustrate that good PE strategies may not always translate to good SDE strategies.

pdf bib abs
Refining App Reviews: Dataset, Methodology, and Evaluation
Amrita Singh | Chirag Jain | Mohit Chaudhary | Preethu Rose Anish

With the growing number of mobile users, app development has become increasingly lucrative. Reviews on platforms such as Google Play and Apple App Store provide valuable insights to developers, highlighting bugs, suggesting new features, and offering feedback. However, many reviews contain typos, spelling errors, grammar mistakes, and complex sentences, hindering efficient interpretation and slowing down app improvement processes. To tackle this, we introduce RARE (Repository for App review REfinement), a benchmark dataset of 10,000 annotated pairs of original and refined reviews from 10 mobile applications. These reviews were collaboratively refined by humans and large language models (LLMs). We also conducted an evaluation of eight state-of-the-art LLMs for automated review refinement. The top-performing model (Flan-T5) was further used to refine an additional 10,000 reviews, contributing to RARE as a silver corpus.

The telecommunications industry, characterized by its vast customer base and complex service offerings, necessitates a high level of domain expertise and proficiency in customer service center operations. Consequently, there is a growing demand for Large Language Models (LLMs) to augment the capabilities of customer service representatives. This paper introduces a methodology for developing a specialized Telecommunications LLM (Telco LLM) designed to enhance the efficiency of customer service agents and promote consistency in service quality across representatives. We present the construction process of TelBench, a novel dataset created for performance evaluation of customer service expertise in the telecommunications domain. We also evaluate various LLMs and demonstrate the ability to benchmark both proprietary and open-source LLMs on predefined telecommunications-related tasks, thereby establishing metrics that define telcommunications performance.

Large Language Models (LLMs) excel at understanding the semantic relationships between queries and documents, even with lengthy and complex long-tail queries. These queries are challenging for feedback-based rankings due to sparse user engagement and limited feedback, making LLMs’ ranking ability highly valuable. However, the large size and slow inference of LLMs necessitate the development of smaller, more efficient models (sLLMs). Recently, integrating ranking label generation into distillation techniques has become crucial, but existing methods underutilize LLMs’ capabilities and are cumbersome. Our research, RRADistill: Re-Ranking Ability Distillation, propose an efficient label generation pipeline and novel sLLM training methods for both encoder and decoder models. We introduce an encoder-based method using a Term Control Layer to capture term matching signals and a decoder-based model with a ranking layer for enhanced understanding. A/B testing on a Korean-based search platform, validates the effectiveness of our approach in improving re-ranking for long-tail queries.

pdf bib abs
KorSmishing Explainer: A Korean-centric LLM-based Framework for Smishing Detection and Explanation Generation
Yunseung Lee | Daehee Han

To mitigate the annual financial losses caused by SMS phishing (smishing) in South Korea, we propose an explainable smishing detection framework that adapts to a Korean-centric large language model (LLM). Our framework not only classifies smishing attempts but also provides clear explanations, enabling users to identify and understand these threats. This end-to-end solution encompasses data collection, pseudo-label generation, and parameter-efficient task adaptation for models with fewer than five billion parameters. Our approach achieves a 15% improvement in accuracy over GPT-4 and generates high-quality explanatory text, as validated by seven automatic metrics and qualitative evaluation, including human assessments.

pdf bib abs
Time Matters: An End-to-End Solution for Temporal Claim Verification
Anab Maulana Barik | Wynne Hsu | Mong-Li Lee

Automated claim verification plays an essential role in fostering trust in the digital space. Despite the growing interest, the verification of temporal claims has not received much attention in the community. Temporal claim verification brings new challenges where cues of the temporal information need to be extracted, and temporal reasoning involving various temporal aspects of the text must be applied.In this work, we describe an end-to-end solution for temporal claim verification that considers the temporal information in claims to obtain relevant evidence sentences and harnesses the power of a large language model for temporal reasoning. We curate two datasets comprising a diverse range of temporal claims to learn time-sensitive representations that encapsulate not only the semantic relationships among the events, but also their chronological proximity.Experiment results demonstrate that the proposed approach significantly enhances the accuracy of temporal claim verification, thereby advancing current state-of-the-art in automated claim verification.

This study introduces a Multidisciplinary chILDhood cancer survivor question-answering (MILD) bot designed to support childhood cancer survivors facing diverse challenges in their survivorship journey. In South Korea, a shortage of experts equipped to address these unique concerns comprehensively leaves survivors with limited access to reliable information. To bridge this gap, our MILD bot employs a dual-component model featuring an intent classifier and a semantic textual similarity model. The intent classifier first analyzes the user’s query to identify the underlying intent and match it with the most suitable expert who can provide advice. Then, the semantic textual similarity model identifies questions in a predefined dataset that closely align with the user’s query, ensuring the delivery of relevant responses. This proposed framework shows significant promise in offering timely, accurate, and high-quality information, effectively addressing a critical need for support among childhood cancer survivors.

Generative retrieval (GR) has emerged as a transformative paradigm in search and recommender systems, leveraging numeric-based identifier representations to enhance efficiency and generalization. Notably, methods like TIGER, which employ Residual Quantization-based Semantic Identifiers (RQ-SID), have shown significant promise in e-commerce scenarios by effectively managing item IDs. However, a critical issue termed the "Hourglass" phenomenon, occurs in RQ-SID, where intermediate codebook tokens become overly concentrated, hindering the full utilization of generative retrieval methods. This paper analyses and addresses this problem by identifying data sparsity and long-tailed distribution as the primary causes. Through comprehensive experiments and detailed ablation studies, we analyze the impact of these factors on codebook utilization and data distribution. Our findings reveal that the “Hourglass” phenomenon substantially impacts the performance of RQ-SID in generative retrieval. We propose effective solutions to mitigate this issue, thereby significantly enhancing the effectiveness of generative retrieval in real-world E-commerce applications.

pdf bib abs
Improving Few-Shot Cross-Domain Named Entity Recognition by Instruction Tuning a Word-Embedding based Retrieval Augmented Large Language Model
Subhadip Nandi | Neeraj Agrawal

Few-Shot Cross-Domain NER is the process of leveraging knowledge from data-rich source domains to perform entity recognition on data-scarce target domains. Most previous state-of-the-art (SOTA) approaches use pre-trained language models (PLMs) for cross-domain NER. However, these models are often domain specific. To successfully use these models for new target domains, we need to modify either the model architecture or perform model fine-tuning using data from the new domains. Both of these result in the creation of entirely new NER models for each target domain which is infeasible for practical scenarios. Recently, several works have attempted to use LLMs to solve Few-Shot Cross-Domain NER. However, most of these are either too expensive for practical purposes or struggle to follow LLM prompt instructions. In this paper, we propose IF-WRANER (Instruction Finetuned Word-embedding based Retrieval Augmented large language model for Named Entity Recognition), a retrieval augmented LLM, finetuned for the NER task. By virtue of the regularization techniques used during LLM finetuning and the adoption of word-level embedding over sentence-level embedding during the retrieval of in-prompt examples, IF-WRANER is able to outperform previous SOTA Few-Shot Cross-Domain NER approaches. We have demonstrated the effectiveness of our model by benchmarking its performance on the open source CrossNER dataset, on which it shows more than 2% F1 score improvement over the previous SOTA model. We have deployed the model for multiple customer care domains of an enterprise. Accurate entity prediction through IF-WRANER helps direct customers to automated workflows for the domains, thereby reducing escalations to human agents by almost 15% and leading to millions of dollars in yearly savings for the company.

Unlike professional Business-to-Consumer (B2C) e-commerce platforms (e.g., Amazon), Consumer-to-Consumer (C2C) platforms (e.g., Facebook marketplace) are mainly targeting individual sellers who usually lack sufficient experience in e-commerce. Individual sellers often struggle to compose proper descriptions for selling products. With the recent advancement of Multimodal Large Language Models (MLLMs), we attempt to integrate such state-of-the-art generative AI technologies into the product listing process. To this end, we develop IPL, an Intelligent Product Listing tool tailored to generate descriptions using various product attributes such as category, brand, color, condition, etc. IPL enables users to compose product descriptions by merely uploading photos of the selling product. More importantly, it can imitate the content style of our C2C platform Xianyu. This is achieved by employing domain-specific instruction tuning on MLLMs, and by adopting the multi-modal Retrieval-Augmented Generation (RAG) process. A comprehensive empirical evaluation demonstrates that the underlying model of IPL significantly outperforms the base model in domain-specific tasks while producing less hallucination. IPL has been successfully deployed in our production system, where 72% of users have their published product listings based on the generated content, and those product listings are shown to have a quality score 5.6% higher than those without AI assistance.

Finetuning large language models requires huge GPU memory, restricting the choice to acquire Larger models. While the quantized version of the Low-Rank Adaptation technique, named QLoRA, significantly alleviates this issue, finding the efficient LoRA rank is still challenging. Moreover, QLoRA is trained on a pre-defined rank and, therefore, cannot be reconfigured for its lower ranks without requiring further fine-tuning steps. This paper proposes QDyLoRA -Quantized Dynamic Low-Rank Adaptation-, as an efficient quantization approach for dynamic low-rank adaptation. Motivated by Dynamic LoRA, QDyLoRA is able to efficiently finetune LLMs on a set of pre-defined LoRA ranks. QDyLoRA enables fine-tuning Falcon-40b for ranks 1 to 64 on a single 32 GB V100-GPU through one round of fine-tuning. Experimental results show that QDyLoRA is competitive to QLoRA and outperforms when employing its optimal rank.

pdf bib abs
Improving Hierarchical Text Clustering with LLM-guided Multi-view Cluster Representation
Anup Pattnaik | Cijo George | Rishabh Kumar Tripathi | Sasanka Vutla | Jithendra Vepa

In this work, we present an approach that introduces different perspectives or views to improve the quality of hierarchical clustering of interaction drivers in a contact center. Specifically, we present a multi-stage approach that introduces LLM-guided multi-view cluster representation that significantly improves the quality of generated clusters. Our approach improves average Silhouette Score by upto 70% and Human Preference Scores by 36.7% for top-level clusters compared to standard agglomerative clustering for the given business use-case. We also present how the proposed approach can be adapted to cater to a standard non-hierarchical clustering use-cases where it achieves state-of-the-art performance on public datasets based on NMI and ACC scores, with minimal number of LLM queries compared to the current state-of-the-art approaches. Moreover, we apply our technique to generate two new labeled datasets for hierarchical clustering. We open-source these labeled datasets, validated and corrected by domain experts, for the benefit of the research community.

pdf bib abs
PARA: Parameter-Efficient Fine-tuning with Prompt-Aware Representation Adjustment
Zequan Liu | Yi Zhao | Ming Tan | Wei Zhu | Aaron Xuxiang Tian

In the realm of parameter-efficient fine-tuning (PEFT) methods, while options like LoRA are available, there is a persistent demand in the industry for a PEFT approach that excels in both efficiency and performance within the context of single-backbone multi-tenant applications. This paper introduces a new and straightforward PEFT technique, termed Prompt Aware Representation Adjustment (PARA). The core of our proposal is to integrate a lightweight vector generator within each Transformer layer. This generator produces vectors that are responsive to input prompts, thereby adjusting the hidden representations accordingly. Our extensive experimentation across diverse tasks has yielded promising results. Firstly, the PARA method has been shown to surpass current PEFT benchmarks in terms of performance, despite having a similar number of adjustable parameters. Secondly, it has proven to be more efficient than LoRA in the single-backbone multi-tenant scenario, highlighting its significant potential for industrial adoption.

With the ever-increasing demands on Question Answering (QA) systems for IT operations and maintenance, an efficient and supervised fine-tunable framework is necessary to ensure the data security, private deployment and continuous upgrading. Although Large Language Models (LLMs) have notably improved the open-domain QA’s performance, how to efficiently handle enterprise-exclusive corpora and build domain-specific QA systems are still less-studied for industrial applications. In this paper, we propose a general and comprehensive framework based on Retrieval Augmented Generation (RAG) and facilitate the whole business process of establishing QA systems for IT operations and maintenance. In accordance with the prevailing RAG method, our proposed framework, named with RAG4ITOps, composes of two major stages: (1) Models Fine-tuning & Data Vectorization, and (2) Online QA System Process. At the Stage 1, we leverage a contrastive learning method with two negative sampling strategies to fine-tune the embedding model, and design the instruction templates to fine-tune the LLM with a Retrieval Augmented Fine-Tuning method. At the Stage 2, an efficient process of QA system is built for serving. We collect enterprise-exclusive corpora from the domain of cloud computing, and the extensive experiments show that our method achieves superior results than counterparts on two kinds of QA tasks. Our experiment also provide a case for applying the RAG4ITOps to real-world enterprise-level applications.

In recent years, large language models (LLMs) have attracted significant interest from the research community due to their broad applicability in many language-oriented tasks, and are now widely used in numerous areas of production and daily life. One source of the powerful capabilities of LLMs is the massive scale of their pre-training dataset. However, these pre-training datasets contain many outdated, harmful, and personally sensitive information, which inevitably becomes memorized by LLM during the pre-training process. Eliminating this undesirable data is crucial for ensuring the model’s safety and enhancing the user experience. However, the cost of extensively cleaning the pre-training dataset and retraining the model from scratch is very high. In this work, we propose ULMR , a unlearning framework for LLMs , which first uses carefully designed prompts to rewrite the instructions in the specified dataset, and generate corresponding negative responses. Subsequently, to ensure that the model does not excessively deviate post-training, we perform model parameter averaging to preserve the performance of the original LLM. We conducted experiments on two public datasets, TOFU and RWKU, demonstrating that our method can effectively forget specified information while retaining the capabilities of the original LLM.

pdf bib abs
Pretraining and Finetuning Language Models on Geospatial Networks for Accurate Address Matching
Saket Maheshwary | Arpan Paul | Saurabh Sohoney

We propose a novel framework for pretraining and fine-tuning language models with the goal of determining whether two addresses represent the same physical building. Address matching and building authoritative address catalogues are important to many applications and businesses, such as delivery services, online retail, emergency services, logistics, etc. We propose to view a collection of addresses as an address graph and curate inputs for language models by placing geospatially linked addresses in the same context. Our approach jointly integrates concepts from graph theory and weak supervision with address text and geospatial semantics. This integration enables us to generate informative and diverse address pairs, facilitating pretraining and fine-tuning in a self-supervised manner. Experiments and ablation studies on manually curated datasets and comparisons with state-of-the-art techniques demonstrate the efficacy of our approach. We achieve a 24.49% improvement in recall while maintaining 95% precision on average, in comparison to the current baseline across multiple geographies. Further, we deploy our proposed approach and show the positive impact of improving address matching on geocode learning.

pdf bib abs
SMARTCAL: An Approach to Self-Aware Tool-Use Evaluation and Calibration
Yuanhao Shen | Xiaodan Zhu | Lei Chen

The tool-use ability of Large Language Models (LLMs) has a profound impact on a wide range of applications. However, LLMs’ self-awareness and self-control capability in appropriately using tools remains understudied. The problem is consequential as it alarms a potential risk of degraded performance and poses a threat to trustworthiness on the models. In this paper, we conduct a study on a family of state-of-the-art LLMs on three datasets with two mainstream tool-use frameworks. Our study reveals the tool-abuse behavior of LLMs, a tendency for models to misuse tools along with models’ frequent overconfidence in tool choice. We also find that this is a common issue regardless of model capability. Accordingly, we propose a novel framework, SMARTCAL, to mitigate the observed issues, and our results show an average 8.6 percent increase in the QA performance in three testing datasets and 21.6 percent lower Expected Calibration Error (ECE) than existing methods.

Recent advancements in large Language Models (LMs) have significantly enhanced their capabilities across various domains, including natural language understanding and generation. In this paper, we investigate the application of LMs to the specialized task of contact-center Quality Assurance (QA), which involves evaluating conversations between human agents and customers. This task requires both sophisticated linguistic understanding and deep domain knowledge. We conduct a comprehensive assessment of eight LMs, revealing that larger models, such as Claude-3.5-Sonnet, exhibit superior performance in comprehending contact-center conversations. We introduce methodologies to transfer this domain-specific knowledge to smaller models by leveraging evaluation plans generated by more knowledgeable models, with optional human-in-the-loop refinement to enhance the capabilities of smaller models. Notably, our experimental results demonstrate an improvement of up to 18.95% in Macro F1 on an in-house QA dataset. Our findings emphasize the importance of evaluation plans in guiding reasoning and highlight the potential of AI-assisted tools to advance objective, consistent, and scalable agent evaluation processes in contact centers.

In this paper, we study the problem of generating structured objects that conform to a complex schema, with intricate dependencies between the different components (facets) of the object. The facets of the object (attributes, fields, columns, properties) can be a mix of short, structured facts, or long natural-language descriptions. The object has to be self-consistent between the different facets in the redundant information it carries (relative consistency), while being grounded with respect to world knowledge (absolute consistency). We frame the problem as a Language Modeling problem (Structured Object Language Modeling) and train an LLM to perform the task natively, without requiring instructions or prompt-engineering. We propose a self-supervised denoising method to train the model from an existing dataset of such objects. The input query can be the existing object itself, in which case the system acts as a regenerator, completing, correcting, normalizing the input, or any unstructured blurb to be structured. We show that the self-supervised denoising training provides a strong baseline, and that additional supervised fine-tuning with small amount of human demonstrations leads to further improvement. Experimental results show that the proposed method matches or outperforms prompt-engineered general-purpose state-of-the-art LLMs (Claude 3, Mixtral-8x7B), while being order-of-magnitude more cost-efficient.

Breastfeeding and Maternity experts are a scarce resource and engaging in a conversation with mothers on such a sensitive topic is a time-consuming effort. We present our journey and rationale in assisting experts to answer queries about Breastfeeding and Maternity topics from users, mainly mothers. We started by developing a RAG approach to response generation where the generated response is made available to the expert who has the option to draft an answer using the generated text or to answer from scratch. This was the start of an ongoing effort to develop a pipeline of AI/NLP-based functionalities to help experts understand user queries and craft their responses.

pdf bib abs
A Hassle-free Algorithm for Strong Differential Privacy in Federated Learning Systems
Hugh Brendan McMahan | Zheng Xu | Yanxiang Zhang

Differential privacy (DP) and federated learning (FL) are combined as advanced privacy-preserving methods when training on-device language models in production mobile keyboard applications. DP-Follow-the-Regularized-Leader (DP-FTRL) algorithms, leveraging correlated noise mechanisms such as tree aggregation or matrix factorization, are widely used in practice for their superior privacy-utility trade-off and compatibility with FL systems. This paper presents a novel variant of DP-FTRL by adapting the recent theoretical advancements of the Buffered Linear Toeplitz (BLT) mechanism to multi-participant scenarios. In the FL setting, our BLT mechanism demonstrates enhanced privacy-utility trade-off and improved memory efficiency than the widely used tree aggregation mechanism. Moreover, BLT achieves comparable privacy and utility to the state-of-the-art banded matrix factorization mechanism, while significantly simplifying usage requirements and reducing memory. The flexibility of the BLT mechanism allows seamless integration with existing DP FL implementations in production environments. We evaluate the BLT-DP-FTRL algorithm on the StackOverflow dataset, serving as a research simulation benchmark, and across four on-device language model tasks in a production FL system. Our empirical results highlight the potential of the BLT mechanism to elevate the practicality and effectiveness of DP in real-world scenarios.

We propose Project Context for Code Summarization with LLMs (ProConSuL), a new framework to provide a large language model (LLM) with precise information about the code structure from program analysis methods such as a compiler or IDE language services and use task decomposition derived from the code structure. ProConSuL builds a call graph to provide the context from callees and uses a two-phase training method (SFT + preference alignment) to train the model to use the project context. We also provide a new evaluation benchmark for C/C++ functions and a set of proxy metrics. Experimental results demonstrate that ProConSuL allows to significantly improve code summaries and reduce the number of hallucinations compared to the base model (CodeLlama-7B-instruct). We make our code and dataset available at https://github.com/TypingCat13/ProConSuL.

pdf bib abs
Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approach
Zhuowan Li | Cheng Li | Mingyang Zhang | Qiaozhu Mei | Michael Bendersky

Retrieval Augmented Generation (RAG) has been a powerful tool for Large Language Models (LLMs) to efficiently process overly lengthy contexts. However, recent LLMs like Gemini-1.5 and GPT-4 show exceptional capabilities to understand long contexts directly. We conduct a comprehensive comparison between RAG and long-context (LC) LLMs, aiming to leverage the strengths of both. We benchmark RAG and LC across various public datasets using three latest LLMs. Results reveal that when resourced sufficiently, LC consistently outperforms RAG in terms of average performance. However, RAG’s significantly lower cost remains a distinct advantage. Based on this observation, we propose Self-Route, a simple yet effective method that routes queries to RAG or LC based on model self-reflection. Self-Route significantly reduces the computation cost while maintaining a comparable performance to LC. Our findings provide a guideline for long-context applications of LLMs using RAG and LC.

pdf bib
MARS: Multilingual Aspect-centric Review Summarisation
Sandeep Sricharan Mukku | Abinesh Kanagarajan | Chetan Aggarwal | Promod Yenigalla

pdf bib abs
A new approach for fine-tuning sentence transformers for intent classification and out-of-scope detection tasks
Tianyi Zhang | Atta Norouzian | Aanchan Mohan | Frederick Ducatelle

In virtual assistant (VA) systems it is important to reject or redirect user queries that fall outside the scope of the system. One of the most accurate approaches for out-of-scope (OOS) rejection is to combine it with the task of intent classification on in-scope queries, and to use methods based on the similarity of embeddings produced by transformer-based sentence encoders. Typically, such encoders are fine-tuned for the intent-classification task, using cross-entropy loss. Recent work has shown that while this produces suitable embeddings for the intent-classification task, it also tends to disperse in-scope embeddings over the full sentence embedding space. This causes the in-scope embeddings to potentially overlap with OOS embeddings, thereby making OOS rejection difficult. This is compounded when OOS data is unknown. To mitigate this issue our work proposes to regularize the cross-entropy loss with an in-scope embedding reconstruction loss learned using an auto-encoder. Our method achieves a 1-4% improvement in the area under the precision-recall curve for rejecting out-of-sample (OOS) instances, without compromising intent classification performance.

pdf bib abs
Tell me what I need to know: Exploring LLM-based (Personalized) Abstractive Multi-Source Meeting Summarization
Frederic Kirstein | Terry Ruas | Robert Kratel | Bela Gipp

Meeting summarization is crucial in digital communication, but existing solutions struggle with salience identification to generate personalized, workable summaries, and context understanding to fully comprehend the meetings’ content.Previous attempts to address these issues by considering related supplementary resources (e.g., presentation slides) alongside transcripts are hindered by models’ limited context sizes and handling the additional complexities of the multi-source tasks, such as identifying relevant information in additional files and seamlessly aligning it with the meeting content.This work explores multi-source meeting summarization considering supplementary materials through a three-stage large language model approach: identifying transcript passages needing additional context, inferring relevant details from supplementary materials and inserting them into the transcript, and generating a summary from this enriched transcript.Our multi-source approach enhances model understanding, increasing summary relevance by ~9% and producing more content-rich outputs.We introduce a personalization protocol that extracts participant characteristics and tailors summaries accordingly, improving informativeness by ~10%.This work further provides insights on performance-cost trade-offs across four leading model families, including edge-device capable options.Our approach can be extended to similar complex generative tasks benefitting from additional resources and personalization, such as dialogue systems and action planning.

The high capability of recent Large Language Models (LLMs) has led to concerns about possible misuse as cheating assistants in open-ended writing tasks in assessments. Although various detecting methods have been proposed, most of them have not been evaluated on or optimized for real-world samples from LLM-assisted cheating, where the generated text is often copy-typed imperfectly by the test-taker. In this paper, we present a framework for training LLM-generated text detectors that can effectively detect LLM-generated samples after being copy-typed. We enhance the existing transformer-based classifier training process with contrastive learning on constructed pairwise data and self-training on unlabeled data, and evaluate the improvements on a real-world dataset from the Duolingo English Test (DET), a high-stakes online English proficiency test. Our experiments demonstrate that the improved model outperforms the original transformer-based classifier and other baselines.

Mitigating bias in language models (LMs) has become a critical problem due to the widespread deployment of LMs in the industry and customer-facing applications. Numerous approaches revolve around data pre-processing and subsequent fine-tuning of language models, tasks that can be both time-consuming and computationally demanding. As alternatives, machine unlearning techniques are being explored, yet there is a notable lack of comparative studies evaluating the effectiveness of these methods. In this work, we explore the effectiveness of two machine unlearning methods: Partitioned Contrastive Gradient Unlearning (PCGU) applied on decoder models, and Negation via Task Vector, and compare them with Direct Preference Optimization (DPO) to reduce social biases in open-source LMs such as LLaMA-2 and OPT. We also implement distributed PCGU for large models. It is empirically shown, through quantitative and qualitative analyses, that negation via Task Vector method outperforms PCGU and is comparable to DPO in debiasing models with minimum deterioration in model performance and perplexity. Negation via Task Vector reduces the bias score by 25.5% for LLaMA-2 and achieves bias reduction of up to 40% for OPT models. Moreover, it can be easily tuned to balance the trade-off between bias reduction and generation quality, unlike DPO.

pdf bib abs
Don’t be my Doctor! Recognizing Healthcare Advice in Large Language Models
Kellen Tan Cheng | Anna Lisa Gentile | Pengyuan Li | Chad DeLuca | Guang-Jie Ren

Large language models (LLMs) have seen increasing popularity in daily use, with their widespread adoption by many corporations as virtual assistants, chatbots, predictors, and many more. Their growing influence raises the need for safeguards and guardrails to ensure that the outputs from LLMs do not mislead or harm users. This is especially true for highly regulated domains such as healthcare, where misleading advice may influence users to unknowingly commit malpractice. Despite this vulnerability, the majority of guardrail benchmarking datasets do not focus enough on medical advice specifically. In this paper, we present the HeAL benchmark (HEalth Advice in LLMs), a health-advice benchmark dataset that has been manually curated and annotated to evaluate LLMs’ capability in recognizing health-advice - which we use to safeguard LLMs deployed in industrial settings. We use HeAL to assess several models and report a detailed analysis of the findings.

pdf bib abs
Building an Efficient Multilingual Non-Profit IR System for the Islamic Domain Leveraging Multiprocessing Design in Rust
Vera Pavlova | Mohammed Makhlouf

The widespread use of large language models (LLMs) has dramatically improved many applications of Natural Language Processing (NLP), including Information Retrieval (IR). However, domains that are not driven by commercial interest often lag behind in benefiting from AI-powered solutions. One such area is religious and heritage corpora. Alongside similar domains, Islamic literature holds significant cultural value and is regularly utilized by scholars and the general public. Navigating this extensive amount of text is challenging, and there is currently no unified resource that allows for easy searching of this data using advanced AI tools. This work focuses on the development of a multilingual non-profit IR system for the Islamic domain. This process brings a few major challenges, such as preparing multilingual domain-specific corpora when data is limited in certain languages, deploying a model on resource-constrained devices, and enabling fast search on a limited budget. By employing methods like continued pre-training for domain adaptation and language reduction to decrease model size, a lightweight multilingual retrieval model was prepared, demonstrating superior performance compared to larger models pre-trained on general domain data. Furthermore, evaluating the proposed architecture that utilizes Rust Language capabilities shows the possibility of implementing efficient semantic search in a low-resource setting.

API integration is crucial for enterprise systems, as it enables seamless interaction between applications within workflows. However, the diversity and complexity of the API landscape present significant challenges in combining API calls based on user intent.Existing methods rely on named entity recognition (NER) and knowledge graphs, but struggle to generate more complex control flow structures, such as conditionals and loops.We propose a novel framework that leverages the success of large language models (LLMs) in code generation to integrate APIs based on natural language input. Our approach involves fine-tuning an LLM using automatically generated API flows derived from OpenAPI specifications.We further evaluate the effectiveness of enforcing the syntax and schema adherence through constrained decoding.To enable systematic comparison, we introduce targeted test suites to assess the generalization capabilities of these approaches and their ability to retain structured knowledge.Our findings show that LLMs fine-tuned on OpenAPI specifications can (a) learn structural API constraints implicitly during training, and (b) achieve significant improvements in both in-distribution and out-of-distribution performance over NER and retrieval-augmented generation (RAG)-based approaches.

We introduce OMG-QA, a new resource for question answering that is designed to evaluate the effectiveness of question answering systems that perform retrieval augmented generation (RAG) in scenarios that demand reasoning on multi-modal, multi-document contexts. These systems, given a user query, must retrieve relevant contexts from the web, which may include non-textual information, and then reason and synthesize these contents to generate a detailed, coherent answer. Unlike existing open-domain QA datasets, OMG-QA requires systems to navigate and integrate diverse modalities and a broad pool of information sources, making it uniquely challenging. We conduct a thorough evaluation and analysis of a diverse set of QA systems, featuring various retrieval frameworks, document retrievers, document indexing approaches, evidence retrieval methods, and LLMs tasked with both information retrieval and generation. Our findings reveal significant limitations in existing approaches using RAG or LLM agents to address open questions that require long-form answers supported by multi-modal evidence. We believe that OMG-QA will be a valuable resource for developing QA systems that are better equipped to handle open-domain, multi-modal information-seeking tasks.

pdf bib abs
Survival of the Safest: Towards Secure Prompt Optimization through Interleaved Multi-Objective Evolution
Ankita Sinha | Wendi Cui | Kamalika Das | Jiaxin Zhang

Large language models (LLMs) have demonstrated remarkable capabilities; however, the optimization of their prompts has historically prioritized performance metrics at the expense of crucial safety and security considerations. To overcome this shortcoming, we introduce “Survival of the Safest” (), an innovative multi-objective prompt optimization framework that enhances both performance and security in LLMs simultaneously. utilizes an interleaved multi-objective evolution strategy, integrating semantic, feedback, and crossover mutations to effectively traverse the prompt landscape. Differing from the computationally demanding Pareto front methods, provides a scalable solution that expedites optimization in complex, high-dimensional discrete search spaces while keeping computational demands low. Our approach accommodates flexible weighting of objectives and generates a pool of optimized candidates, empowering users to select prompts that optimally meet their specific performance and security needs. Experimental evaluations across diverse benchmark datasets affirm ‘s efficacy in delivering high performance and notably enhancing safety and security compared to single-objective methods. This advancement marks a significant stride towards the deployment of LLM systems that are both high-performing and secure across varied industrial applications

pdf bib abs
Fine-Tuning Large Language Models for Stock Return Prediction Using Newsflow
Tian Guo | Emmanuel Hauptmann

Large language models (LLMs) and their fine-tuning techniques have demonstrated superior performance in various language understanding and generation tasks.This paper explores fine-tuning LLMs for predicting stock returns with financial newsflow.Return prediction is fundamental for subsequent tasks like portfolio construction and optimization in quantitative investing. We formulate the model to include a text representation and forecasting modules. We propose to compare the encoder-only and decoder-only LLMs, considering they generate text representations in distinct ways.The impact of these different representations on return forecasting remains an open question.Meanwhile, we compare two simple methods of integrating LLMs’ token-level representations into the forecasting module.The experiments on real investment universes reveal that:(1) aggregated representations from LLMs’ token-level embeddings generally produce return predictions that enhance the performance of long-only and long-short portfolios;(2) in the relatively large investment universe, the decoder LLMs-based prediction model leads to stronger portfolios, whereas in the small universes, there are no consistent winners;(3) return predictions derived from LLMs’ text representations are a strong signal for portfolio construction, outperforming conventional sentiment scores.These findings shed light on developing suitable LLM fine-tuning methods for return prediction-based portfolio construction.

pdf bib abs
AmazonQAC: A Large-Scale, Naturalistic Query Autocomplete Dataset
Dante Everaert | Rohit Patki | Tianqi Zheng | Christopher Potts

Query Autocomplete (QAC) is a critical feature in modern search engines, facilitating user interaction by predicting search queries based on input prefixes. Despite its widespread adoption, the absence of large-scale, realistic datasets has hindered advancements in QAC system development. This paper addresses this gap by introducing AmazonQAC, a new QAC dataset sourced from Amazon Search logs, comprising 395M samples. The dataset includes actual sequences of user-typed prefixes leading to final search terms, as well as session IDs and timestamps that support modeling the context-dependent aspects of QAC. We assess Prefix Trees, semantic retrieval, and Large Language Models (LLMs) with and without finetuning. We find that finetuned LLMs perform best, particularly when incorporating contextual information. However, even our best system achieves only half of what we calculate is theoretically possible on our test data, which implies QAC is a challenging problem that is far from solved with existing systems. This contribution aims to stimulate further research on QAC systems to better serve user needs in diverse environments. We open-source this data on Hugging Face at https://huggingface.co/datasets/amazon/AmazonQAC.

This paper presents LOFI (Language, OCR, Form Independent), a pipeline for Document Information Extraction (DIE) in Low-Resource Language (LRL) business documents. LOFI pipeline solves language, Optical Character Recognition (OCR), and form dependencies through flexible model architecture, a token-level box split algorithm, and the SPADE decoder. Experiments on Korean and Japanese documents demonstrate high performance in Semantic Entity Recognition (SER) task without additional pre-training. The pipeline’s effectiveness is validated through real-world applications in insurance and tax-free declaration services, advancing DIE capabilities for diverse languages and document types in industrial settings.

The Chartered Financial Analyst (CFA) program is one of the most widely recognized financial certifications globally. In this work, we test a variety of state-of-the-art large language models (LLMs) on mock CFA exams to provide an overview of their financial analysis capabilities using the same evaluation standards applied for human professionals. We benchmark five leading proprietary models and eight open-source models on all three levels of the CFA through challenging multiple-choice and essay questions. We find that flagship proprietary models perform relatively well and can solidly pass levels I and II exams, but fail at level III due to essay questions. Open-source models generally fall short of estimated passing scores, but still show strong performance considering their size, cost, and availability advantages. We also find that using textbook data helps bridge the gap between open-source and proprietary models to a certain extent, despite reduced gains in CFA levels II and III. By understanding the current financial analysis abilities of LLMs, we aim to guide practitioners on which models are best suited for enhancing automation in the financial industry.

Aligning large language models (LLMs) to value systems has emerged as a significant area of research within the fields of AI and NLP. Currently, this alignment process relies on the availability of high-quality supervised and preference data, which can be both time-consuming and expensive to curate or annotate. In this paper, we introduce a systematic end-to-end methodology for aligning LLMs to the implicit and explicit values represented in unstructured text data. Our proposed approach leverages the use of scalable synthetic data generation techniques to effectively align the model to the values present in the unstructured data. Through two distinct use-cases, we demonstrate the efficiency of our methodology on the Mistral-7B-Instruct model. Our approach credibly aligns LLMs to the values embedded within documents, and shows improved performance against other approaches, as quantified through the use of automatic metrics and win rates.

pdf bib abs
LARA: Linguistic-Adaptive Retrieval-Augmentation for Multi-Turn Intent Classification
Junhua Liu | Tan Yong Keat | Bin Fu | Kwan Hui Lim

Multi-turn intent classification is notably challenging due to the complexity and evolving nature of conversational contexts. This paper introduces LARA, a Linguistic-Adaptive Retrieval-Augmentation framework to enhance accuracy in multi-turn classification tasks across six languages, accommodating numerous intents in chatbot interactions. LARA combines a fine-tuned smaller model with a retrieval-augmented mechanism, integrated within the architecture of LLMs. The integration allows LARA to dynamically utilize past dialogues and relevant intents, thereby improving the understanding of the context. Furthermore, our adaptive retrieval techniques bolster the cross-lingual capabilities of LLMs without extensive retraining and fine-tuning. Comprehensive experiments demonstrate that LARA achieves state-of-the-art performance on multi-turn intent classification tasks, enhancing the average accuracy by 3.67% from state-of-the-art single-turn intent classifiers.

To enhance a question-answering system for automotive drivers, we tackle the problem of automatic generation of icon image descriptions. The descriptions can match the driver’s query about the icon appearing on the dashboard and tell the driver what is happening so that they may take an appropriate action. We use three state-of-the-art large vision-language models to generate both visual and functional descriptions based on the icon image and its context information in the car manual. Both zero-shot and few-shot prompts are used. We create a dataset containing over 400 icons with their ground-truth descriptions and use it to evaluate model-generated descriptions across several performance metrics. Our evaluation shows that two of these models (GPT-4o and Claude 3.5) performed well on this task, while the third model (LLaVA-NEXT) performs poorly.

pdf bib abs
Athena: Safe Autonomous Agents with Verbal Contrastive Learning
Tanmana Sadhu | Ali Pesaranghader | Yanan Chen | Dong Hoon Yi

Due to emergent capabilities, large language models (LLMs) have been utilized as language-based agents to perform a variety of tasks and make decisions with an increasing degree of autonomy. These autonomous agents can understand high-level instructions, interact with their environments, and execute complex tasks using a selection of tools available to them. As the capabilities of the agents expand, ensuring their safety and trustworthiness becomes more imperative. In this study, we introduce the Athena framework which leverages the concept of verbal contrastive learning where past safe and unsafe trajectories are used as in-context (contrastive) examples to guide the agent towards safety while fulfilling a given task. The framework also incorporates a critiquing mechanism to guide the agent to prevent risky actions at every step. Furthermore, due to the lack of existing benchmarks on the safety reasoning ability of LLM-based agents, we curate a set of 80 toolkits across 8 categories with 180 scenarios to provide a safety evaluation benchmark. Our experimental evaluation, with both closed- and open-source LLMs, indicates verbal contrastive learning and interaction-level critiquing improve the safety rate significantly.

An emergent research trend explores the use of Large Language Models (LLMs) as the backbone of agentic systems (e.g., SWE-Bench, Agent-Bench). To fulfill LLMs’ potential as autonomous agents, they must be able to identify, call, and interact with a variety of external tools and application program interfaces (APIs). This capability of LLMs, commonly termed function calling, leads to a myriad of advantages such as access to current and domain-specific information in databases and the outsourcing of tasks that can be reliably performed by tools. In this work, we introduce Granite-20B-FunctionCalling, a model trained using a multi-task training approach on seven fundamental tasks encompassed in function calling. Our comprehensive evaluation on multiple out-of-domain datasets, which compares Granite-20B-FunctionCalling to more than 15 other best proprietary and open models, shows that Granite-20B-FunctionCalling has better generalizability on multiple tasks across seven different evaluation benchmarks. Moreover, Granite-20B-FunctionCalling shows the best performance among all open models and ranks among the top on the Berkeley Function Calling Leaderboard (BFCL).

pdf bib abs
Query-OPT: Optimizing Inference of Large Language Models via Multi-Query Instructions in Meeting Summarization
Md Tahmid Rahman Laskar | Elena Khasanova | Xue-Yong Fu | Cheng Chen | Shashi Bhushan Tn

This work focuses on the task of query-based meeting summarization in which the summary of a context (meeting transcript) is generated in response to a specific query. When using Large Language Models (LLMs) for this task, a new call to the LLM inference endpoint/API is required for each new query even if the context stays the same. However, repeated calls to the LLM inference endpoints would significantly increase the costs of using them in production, making LLMs impractical for many real-world use cases. To address this problem, in this paper, we investigate whether combining the queries for the same input context in a single prompt to minimize repeated calls can be successfully used in meeting summarization. In this regard, we conduct extensive experiments by comparing the performance of various popular LLMs: GPT-4, Gemini, Claude-3, LLaMA2, Mistral, Phi-3, and Qwen-2 in single-query and multi-query settings. We observe that the capability to reliably generate the response in the expected format is usually limited to closedsource LLMs, with most open-source LLMs lagging behind (except Mistral). We conclude that multi-query prompting could be useful to optimize the inference costs by significantly reducing calls to the inference endpoints/APIs for the task of meeting summarization.

pdf bib abs
DiAL : Diversity Aware Listwise Ranking for Query Auto-Complete
Sonali Singh | Sachin Sudhakar Farfade | Prakash Mandayam Comar

Query Auto-Complete (QAC) is an essential search feature that suggests users with a list of potential search keyword completions as they type, enabling them to complete their queries faster. While the QAC systems in eCommerce stores generally use the Learning to Rank (LTR) approach optimized based on customer feedback, it struggles to provide diverse suggestions, leading to repetitive queries and limited navigational suggestions related to product categories, attributes, and brands. This paper proposes a novel DiAL framework that explicitly optimizes for diversity alongside customer feedback signals. It achieves this by leveraging a smooth approximation of the diversity-based metric (𝛼NDCG) as a listwise loss function and modifying it to balance relevance and diversity. The proposed approach yielded an improvement of 8.5% in mean reciprocal rank (MRR) and 22.8% in 𝛼NDCG compared to the pairwise ranking approach on an eCommerce dataset, while meeting the ultra-low latency constraints of QAC systems. In an online experiment, the diversity-aware listwise QAC model resulted in a 0.48% lift in revenue. Furthermore, we replicated the proposed approach on a publicly available search log, demonstrating improvements in both diversity and relevance of the suggested queries.

pdf bib abs
Systematic Evaluation of Long-Context LLMs on Financial Concepts
Lavanya Gupta | Saket Sharma | Yiyun Zhao

Long-context large language models (LC LLMs) promise to increase reliability of LLMs in real-world tasks requiring processing and understanding of long input documents. However, this ability of LC LLMs to reliably utilize their growing context windows remains under investigation. In this work, we evaluate the performance of state-of-the-art GPT-4 suite of LC LLMs in solving a series of progressively challenging tasks, as a function of factors such as context length, task difficulty, and position of key information by creating a real world financial news dataset. Our findings indicate that LC LLMs exhibit brittleness at longer context lengths even for simple tasks, with performance deteriorating sharply as task complexity increases. At longer context lengths, these state-of-the-art models experience catastrophic failures in instruction following resulting in degenerate outputs. Our prompt ablations also reveal unfortunate continued sensitivity to both the placement of the task instruction in the context window as well as minor markdown formatting. Finally, we advocate for more rigorous evaluation of LC LLMs by employing holistic metrics such as F1 (rather than recall) and reporting confidence intervals, thereby ensuring robust and conclusive findings.

The rapid evolution of Large Language Models (LLMs) and conversational assistants necessitates dynamic, scalable, and configurable conversational datasets for training and evaluation.These datasets must accommodate diverse user interaction modes, including text and voice, each presenting unique modeling challenges. Knowledge Graphs (KGs), with their structured and evolving nature, offer an ideal foundation for current and precise knowledge.Although human-curated KG-based conversational datasets exist, they struggle to keep pace with the rapidly changing user information needs.We present ConvKGYarn, a scalable method for generating up-to-date and configurable conversational KGQA datasets. Qualitative psychometric analyses demonstrate ConvKGYarn’s effectiveness in producing high-quality data comparable to popular conversational KGQA datasets across various metrics.ConvKGYarn excels in adhering to human interaction configurations and operating at a significantly larger scale.We showcase ConvKGYarn’s utility by testing LLMs on diverse conversations — exploring model behavior on conversational KGQA sets with different configurations grounded in the same KG fact set.Our results highlight the ability of ConvKGYarn to improve KGQA foundations and evaluate parametric knowledge of LLMs, thus offering a robust solution to the constantly evolving landscape of conversational assistants.

Crafting a convincing financial market analysis report necessitates a wealth of market information and the expertise of financial analysts, posing a highly challenging task. While large language models (LLMs) have enabled the automated generation of financial market analysis text, they still face issues such as hallucinations, errors in financial knowledge, and insufficient capability to reason about complex financial problems, which limits the quality of the generation. To tackle these shortcomings, we propose a novel task and a retrieval-augmented framework grounded in a financial knowledge graph (FKG). The proposed framework is compatible with commonly used instruction-tuning methods. Experiments demonstrate that our framework, coupled with a small-scale language model fine-tuned with instructions, can significantly enhance the logical consistency and quality of the generated analysis texts, outperforming both large-scale language models and other retrieval-augmented baselines.

Structured generation, the process of producing content in standardized formats like JSON and XML, is widely utilized in real-world applications to extract key output information from large language models (LLMs).This study investigates whether such constraints on generation space impact LLMs’ abilities, including reasoning and domain knowledge comprehension. Specifically, we evaluate LLMs’ performance when restricted to adhere to structured formats versus generating free-form responses across various common tasks. Surprisingly, we observe a significant decline in LLMs’ reasoning abilities under format restrictions. Furthermore, we find that stricter format constraints generally lead to greater performance degradation in reasoning tasks.

pdf bib abs
ASTRA: Automatic Schema Matching using Machine Translation
Tarang Chugh | Deepak Zambre

Many eCommerce platforms source product information from millions of sellers and manufactures, each having their own proprietary schemas, and employ schema matching solutions to structure it to enable informative shopping experiences. Meanwhile, state-of-the-art machine translation techniques have demonstrated great success in building context-aware representations that generalize well to new languages with minimal training data. In this work, we propose modeling the schema matching problem as a neural machine translation task: given product context and an attribute-value pair from a source schema, the model predicts the corresponding attribute, if available, in the target schema. We utilize open-source seq2seq models, such as mT5 and mBART, fine-tuned on product attribute mappings to build a scalable schema matching framework. We demonstrate that our proposed approach achieves a significant performance boost (15% precision and 7% recall uplift) compared to the baseline system and can support new attributes with precision ≥ 95% using only five labeled samples per attribute.

Gboard Decoder produces suggestions by looking for paths that best match input touch points on the context aware search space, which is backed by the language Finite State Transducers (FST). The language FST is currently an N-gram language model (LM). However, N-gram LMs, limited in context length, are known to have sparsity problem under device model size constraint. In this paper, we propose Neural Search Space which substitutes the N-gram LM with a Neural Network LM (NN-LM) and dynamically constructs the search space during decoding. Specifically, we integrate the long range context awareness of NN-LM into the search space by converting its outputs given context, into the language FST at runtime. This involves language FST structure redesign, pruning strategies tuning, and data structure optimizations. Online experiments demonstrate improved quality results, reducing Words Modified Ratio by [0.26%, 1.19%] on various locales with acceptable latency increases. This work opens new avenues for further improving keyboard decoding quality by enhancing neural LM more directly.

Prompt leakage poses a compelling security and privacy threat in LLM applications. Leakage of system prompts may compromise intellectual property, and act as adversarial reconnaissance for an attacker. A systematic evaluation of prompt leakage threats and mitigation strategies is lacking, especially for multi-turn LLM interactions. In this paper, we systematically investigate LLM vulnerabilities against prompt leakage for 10 closed- and open-source LLMs, across four domains. We design a unique threat model which leverages the LLM sycophancy effect and elevates the average attack success rate (ASR) from 17.7% to 86.2% in a multi-turn setting. Our standardized setup further allows dissecting leakage of specific prompt contents such as task instructions and knowledge documents. We measure the mitigation effect of 7 black-box defense strategies, along with finetuning an open-source model to defend against leakage attempts. We present different combination of defenses against our threat model, including a cost analysis. Our study highlights key takeaways for building secure LLM applications and provides directions for research in multi-turn LLM interactions.

The fashion industry is one of the leading domains in the global e-commerce sector, prompting major online retailers to employ recommendation systems for product suggestions and customer convenience. While recommendation systems have been widely studied, most are designed for general e-commerce problems and struggle with the unique challenges of the fashion domain. To address these issues, we propose a sequential fashion recommendation framework that leverages a pre-trained large language model (LLM) enhanced with recommendation-specific prompts. Our framework employs parameter-efficient fine-tuning with extensive fashion data and introduces a novel mix-up-based retrieval technique for translating text into relevant product suggestions. Extensive experiments show our proposed framework significantly enhances fashion recommendation performance.

pdf bib abs
Visual Editing with LLM-based Tool Chaining: An Efficient Distillation Approach for Real-Time Applications
Oren Sultan | Alexander Khasin | Guy Shiran | Asnat Greenstein-Messica | Dafna Shahaf

We present a practical distillation approach to fine-tune LLMs for invoking tools in real-time applications. We focus on visual editing tasks; specifically, we modify images and videos by interpreting user stylistic requests, specified in natural language (“golden hour”), using an LLM to select the appropriate tools and their parameters to achieve the desired visual effect.We found that proprietary LLMs such as GPT-3.5-Turbo show potential in this task, but their high cost and latency make them unsuitable for real-time applications.In our approach, we fine-tune a (smaller) student LLM with guidance from a (larger) teacher LLM and behavioral signals.We introduce offline metrics to evaluate student LLMs. Both online and offline experiments show that our student models manage to match the performance of our teacher model (GPT-3.5-Turbo), significantly reducing costs and latency.Lastly, we show that fine-tuning was improved by 25% in low-data regimes using augmentation.

pdf bib abs
Provenance: A Light-weight Fact-checker for Retrieval Augmented LLM Generation Output
Hithesh Sankararaman | Mohammed Nasheed Yasin | Tanner Sorensen | Alessandro Di Bari | Andreas Stolcke

We present a light-weight approach for detecting nonfactual outputs from retrieval-augemented generation (RAG). Given a context and putative output, we compute a factuality score that can be thresholded to yield a binary decision to check the results of LLM-based question-answering, summarization, or other systems. Unlike factuality checkers that themselves rely on LLMs, we use compact, open-source natural language inference (NLI) models that yield a freely accessible solution with low latency and low cost at run-time, and no need for LLM fine-tuning. The approach also enables downstream mitigation and correction of hallucinations, by tracing them back to specific context chunks. Our experiments show high ROC-AUC across a wide range of relevant open source datasets, indicating the effectiveness of our method for fact-checking RAG output.

We present Any-Modality Augmented Language Model (AnyMAL), a unified model that reasons over diverse input modality signals (i.e. text, image, video, audio, IMU motion sensor), and generates textual responses. AnyMAL inherits the powerful text-based reasoning abilities of the state-of-the-art LLMs including Llama-3 (70B), and converts modality-specific signals to the joint textual space through a pre-trained aligner module.In this paper, we provide details on the optimizations implemented to efficiently scale the training pipeline, and present a comprehensive recipe for model and training configurations. We conduct comprehensive empirical analysis comprising both human and automatic evaluations, and demonstrate state-of-the-art performance on various multimodal tasks compared to industry-leading models – albeit with a relatively small number of trainable parameters.

Most prior safety research of large language models (LLMs) has focused on enhancing the alignment of LLMs to better suit the safety requirements of their use cases. However, internalizing such safeguard features into larger models brought challenges of higher training cost and unintended degradation of helpfulness. In this paper, we leverage a smaller LLM for both harmful query detection and safeguard response generation. We introduce our safety requirements and the taxonomy of harmfulness categories, and then propose a multi-task learning mechanism fusing the two tasks into a single model. We demonstrate the effectiveness of our approach, providing on par or surpassing harmful query detection and safeguard response performance compared to the publicly available LLMs.

pdf bib abs
Hyper-QKSG: Framework for Automating Query Generation and Knowledge-Snippet Extraction from Tables and Lists
Dooyoung Kim | Yoonjin Jang | Dongwook Shin | Chanhoon Park | Youngjoong Ko

These days, there is an increasing necessity to provide a user with a short knowledge-snippet for a query in commercial information retrieval services such as the featured snippet of Google. In this paper, we focus on how to automatically extract the candidates of query-knowledge snippet pairs from structured HTML documents by using a new Language Model (HTML-PLM). In particular, the proposed system is powerful on extracting them from Tables and Lists, and provides a new framework for automate query generation and knowledge-snippet extraction based on a QA-pair filtering procedure including the snippet refinement and verification processes, which enhance the quality of generated query-knowledge snippet pairs. As a result, 53.8% of the generated knowledge-snippets includes complex HTML structures such as tables and lists in our experiments of a real-world environments, and 66.5% of the knowledge-snippets are evaluated as valid.

pdf bib abs
Patentformer: A Novel Method to Automate the Generation of Patent Applications
Juanyan Wang | Sai Krishna Reddy Mudhiganti | Manali Sharma

In recent years, Large Language Models (LLMs) have demonstrated impressive performances across various NLP tasks. However, their potential for automating the task of writing patent documents remains relatively unexplored. To address this gap, in this work, we propose a novel method, Patentformer, for generating patent specification by fine-tuning the generative models with diverse sources of information, e.g., patent claims, drawing text, and brief descriptions of the drawings. To enhance the generative models’ comprehension of the complex task of writing patent specification, we introduce a new task, claim+drawing-to-specification, and release a new dataset. We evaluate our proposed method on thousands of patents from the USPTO and show that our method can generate human-like patent specification in legal writing style. Human evaluations by four patent experts further affirm that our proposed method has the potential to generate correct specification, and the quality of generated specification may sometimes be better than the actual specification.

Large language model advancements have enabled the development of multi-agent frameworks to tackle complex, real-world problems such as to automate workflows that require interactions with diverse tools, reasoning, and human collaboration. We present MARCO, a Multi-Agent Real-time Chat Orchestration framework for automating workflows using LLMs. MARCO addresses key challenges in utilizing LLMs for complex, multi-step task execution in a production environment. It incorporates robust guardrails to steer LLM behavior, validate outputs, and recover from errors that stem from inconsistent output formatting, function and parameter hallucination, and lack of domain knowledge. Through extensive experiments we demonstrate MARCO’s superior performance with 94.48% and 92.74% accuracy on task execution for Digital Restaurant Service Platform conversations and Retail conversations datasets respectively along with 44.91% improved latency and 33.71% cost reduction in a production setting. We also report effects of guardrails in performance gain along with comparisons of various LLM models, both open-source and proprietary. The modular and generic design of MARCO allows it to be adapted for automating workflows across domains and to execute complex tasks through multi-turn interactions.

We present systematic efforts in building long-context multilingual text representation model (TRM) and reranker from scratch for text retrieval. We first introduce a text encoder (base size) enhanced with RoPE and unpadding, pre-trained in a native 8192-token context (longer than 512 of previous multilingual encoders). Then we construct a hybrid TRM and a cross-encoder reranker by contrastive learning. Evaluations show that our text encoder outperforms the same-sized previous state-of-the-art XLM-R. Meanwhile, our TRM and reranker match the performance of large-sized state-of-the-art BGE-M3 models and achieve better results on long-context retrieval benchmarks. Further analysis demonstrate that our proposed models exhibit higher efficiency during both training and inference. We believe their efficiency and effectiveness could benefit various researches and industrial applications.

Citywalk, a recently popular form of urban travel, requires genuine personalization and understanding of fine-grained requests compared to traditional itinerary planning. In this paper, we introduce the novel task of Open-domain Urban Itinerary Planning (OUIP), which generates personalized urban itineraries from user requests in natural language. We then present ItiNera, an OUIP system that integrates spatial optimization with large language models to provide customized urban itineraries based on user needs. This involves decomposing user requests, selecting candidate points of interest (POIs), ordering the POIs based on cluster-aware spatial optimization, and generating the itinerary. Experiments on real-world datasets and the performance of the deployed system demonstrate our system’s capacity to deliver personalized and spatially coherent itineraries compared to current solutions. Source codes of ItiNera are available at https://github.com/YihongT/ITINERA.

pdf bib abs
RESTful-Llama: Connecting User Queries to RESTful APIs
Han Xu | Ruining Zhao | Jindong Wang | Haipeng Chen

Recent advancements in Large Language Models (LLMs) have showcased exceptional performance in zero-shot learning and reasoning tasks. However, integrating these models with external tools - a crucial need for real-world applications - remains a significant challenge. We propose RESTful-Llama, a novel framework designed to enable Llama 3.1 to transform natural language instructions into effective RESTful API calls. To enhance the fine-tuning process, we introduce DOC_Mine, a method to generate fine-tuning datasets from public API documentation. RESTful-Llama distinguishes itself by enabling open-source LLMs to efficiently interact with and adapt to any REST API system. Experiments demonstrate a 31.9% improvement in robustness and a 2.33x increase in efficiency compared to existing methods.

Extracting product information is crucial for informed business decisions and strategic planning across multiple industries. However, recent methods relying only on large language models (LLMs) are resource-intensive and computationally prohibitive due to website structure differences and numerous non-product pages. To address these challenges, we propose a novel modular method that leverages low-cost classification models to filter out company web pages, significantly reducing computational costs. Our approach consists of three modules: web page crawling, product page classification using efficient machine learning models, and product information extraction using LLMs on classified product pages. We evaluate our method on a new dataset of about 7000 product and non-product web pages, achieving a 6-point improvement in F1-score, 95% reduction in computational time, and 87.5% reduction in cost compared to end-to-end LLMs. Our research demonstrates the effectiveness of our proposed low-cost classification module to identify web pages containing product information, making product information extraction more effective and cost-efficient.

Character-based dialogue (CharacterDial) has become essential in the industry (e.g., Character.AI), enabling users to freely customize social characters for social interactions. However, the generalizability and adaptability across various conversational scenarios inherent in customizing social characters still lack public industrial solutions. To address these challenges, by dissecting well-rounded social characters composed of both inherent social profiles and external social behaviors, we manually collect a large-scale Chinese corpus featuring characters with diverse categories and behaviors, and develop CharacterGLM models alongside well-designed refinement methods. Extensive experiments show that CharacterGLM outperforms most popular open- and closed-source LLMs and performs comparably to GPT-4. We will release our data and models for local development and deployment.

pdf bib abs
RAC: Retrieval-augmented Conversation Dataset for Open-domain Question Answering in Conversational Settings
Bonggeun Choi | JeongJae Park | Yoonsung Kim | Jaehyun Park | Youngjoong Ko

In recent years, significant advancements in conversational question and answering (CQA) have been driven by the exponential growth of large language models and the integration of retrieval mechanisms that leverage external knowledge to generate accurate and contextually relevant responses. Consequently, the fields of conversational search and retrieval-augmented generation (RAG) have obtained substantial attention for their capacity to address two key challenges: query rewriting within conversational histories for better retrieval performance and generating responses by employing retrieved knowledge. However, both fields are often independently studied, and comprehensive study on entire systems remains underexplored. In this work, we present a novel retrieval-augmented conversation (RAC) dataset and develop a baseline system comprising query rewriting, retrieval, reranking, and response generation stages. Experimental results demonstrate the competitiveness of the system and extensive analyses are conducted to apprehend the impact of retrieval results to response generation.

pdf bib abs
Improving Retrieval in Sponsored Search by Leveraging Query Context Signals
Akash Kumar Mohankumar | Gururaj K | Gagan Madan | Amit Singh

Accurately retrieving relevant bid keywords for user queries is critical in Sponsored Search but remains challenging, particularly for short, ambiguous queries. Existing dense and generative retrieval models often fail to capture the nuanced user intent in these cases. To address this, we propose an approach to enhance query understanding by augmenting queries with rich contextual signals derived from web search results and large language models, stored in an online cache. Specifically, we use web search titles and snippets to ground queries in real-world information, and utilize GPT-4 to generate query rewrites and explanations that clarify user intent. These signals are efficiently integrated through a Fusion-in-Decoder based Unity architecture, enabling both dense and generative retrieval with serving costs on par with traditional context-free models. To address scenarios where context is unavailable in the cache, we introduce context glancing, a curriculum learning strategy that improves model robustness and performance even without contextual signals during inference. Extensive offline experiments demonstrate that our context-aware approach substantially outperforms context-free models. Furthermore, online A/B testing on a prominent search engine across 160+ countries shows significant improvements in user engagement and revenue.

Large language models (LLMs) have demonstrated prowess in a wide range of tasks. However, many LLMs exhibit significant performance discrepancies between high- and low-resource languages. To mitigate this challenge, we present FuxiTranyu, an open-source multilingual LLM, which is designed to satisfy the need of the research community for balanced and high-performing multilingual capabilities. The base model, FuxiTranyu-8B, features 8 billion parameters and is trained from scratch on meticulously balanced multilingual data that contains 600 billion tokens covering 43 natural languages and 16 programming languages. We also develop two instruction-tuned models: FuxiTranyu-8B-SFT which is fine-tuned on a diverse multilingual instruction dataset, and FuxiTranyu-8B-DPO which is further refined with DPO on a preference dataset for enhanced alignment ability. Extensive experiments on a wide range of multilingual benchmarks demonstrate the competitive performance of FuxiTranyu against existing multilingual LLMs, e.g., BLOOM-7B, PolyLM-13B, and Mistral-7B-Instruct. Both neuron and representation interpretability analyses reveal that FuxiTranyu achieves consistent multilingual representations across languages. To promote further research into multilingual LLMs, we release both the base and instruction-tuned FuxiTranyu models together with 58 pre-training checkpoints at HuggingFace and Github.

pdf bib abs
QUIS: Question-guided Insights Generation for Automated Exploratory Data Analysis
Abhijit Manatkar | Ashlesha Akella | Parthivi Gupta | Krishnasuri Narayanam

Discovering meaningful insights from a large dataset, known as Exploratory Data Analysis (EDA), is a challenging task that requires thorough exploration and analysis of the data. Automated Data Exploration (ADE) systems use goal-oriented methods with Large Language Models and Reinforcement Learning towards full automation. However, these methods require human involvement to anticipate goals that may limit insight extraction, while fully automated systems demand significant computational resources and retraining for new datasets. We introduce QUIS, a fully automated EDA system that operates in two stages: insight generation (ISGen) driven by question generation (QUGen). The QUGen module generates questions in iterations, refining them from previous iterations to enhance coverage without human intervention or manually curated examples. The ISGen module analyzes data to produce multiple relevant insights in response to each question, requiring no prior training and enabling QUIS to adapt to new datasets.

pdf bib abs
PEARL: Preference Extraction with Exemplar Augmentation and Retrieval with LLM Agents
Vijit Malik | Akshay Jagatap | Vinayak S Puranik | Anirban Majumder

Identifying preferences of customers in their shopping journey is a pivotal aspect in providing product recommendations. The task becomes increasingly challenging when there is a multi-turn conversation between the user and a shopping assistant chatbot. In this paper, we tackle a novel and complex problem of identifying customer preferences in the form of key-value filters on an e-commerce website in a multi-turn conversational setting. Existing systems specialize in extracting customer preferences from standalone customer queries which makes them unsuitable to multi-turn setup. We propose PEARL (Preference Extraction with ICL Augmentation and Retrieval with LLM Agents) that leverages collaborative LLM agents, generates in-context learning exemplars and dynamically retrieves relevant exemplars during inference time to extract customer preferences as a combination of key-value filters. Our experiments on proprietary and public datasets show that PEARL not only improves performance on exact match by ~10% compared to competitive LLM-based baselines but additionally improves inference latency by ~110%.

Retrieval-augmented generation (RAG) has emerged as a significant advancement in the field of large language models (LLMs). By integrating up-to-date information not available during their initial training, RAG greatly enhances the practical utility of LLMs in real-world applications. However, even with RAG, LLMs can still produce inaccurate outputs, such as distorting or misinterpreting source content, posing risks in high-trust scenarios. To address these issues, we introduce a novel approach called Hallucination Aware Tuning (HAT). This method involves training hallucination detection models that generate detection labels and provide detailed descriptions of the detected hallucinations. Utilizing these detection results—particularly the hallucination descriptions—GPT-4 Turbo is employed to correct any detected hallucinations. The corrected outputs, free of hallucinations, along with the original versions, are used to create a preference dataset for Direct Preference Optimization (DPO) training. The fine-tuning through DPO leads to LLMs that exhibit a reduced rate of hallucinations and deliver improved answer quality.

pdf bib abs
Intent Detection in the Age of LLMs
Gaurav Arora | Shreya Jain | Srujana Merugu

Intent detection is a critical component of task-oriented dialogue systems (TODS) which enables the identification of suitable actions to address user utterances at each dialog turn. Traditional approaches relied on computationally efficient supervised sentence transformer encoder models, which require substantial training data and struggle with out-of-scope (OOS) detection. The emergence of generative large language models (LLMs) with intrinsic world knowledge presents new opportunities to address these challenges.In this work, we adapt SOTA LLMs using adaptive in-context learning and chain-of-thought prompting for intent detection, and compare their performance with contrastively fine-tuned sentence transformer (SetFit) models to highlight prediction quality and latency tradeoff. We propose a hybrid system using uncertainty based routing strategy to combine the two approaches that along with negative data augmentation results in achieving the best of both worlds ( i.e. within 2% of native LLM accuracy with 50% less latency). To better understand LLM OOS detection capabilities, we perform controlled experiments revealing that this capability is significantly influenced by the scope of intent labels and the size of the label space. We also introduce a two-step approach utilizing internal LLM representations, demonstrating empirical gains in OOS detection accuracy and F1-score by >5% for the Mistral-7B model.

Functional safety is a critical aspect of automotive engineering, encompassing all phases of a vehicle’s lifecycle, including design, development, production, operation, and decommissioning. This domain involves highly knowledge-intensive tasks. This paper introduces Aegis: An Advanced LLM-Based Multi-Agent for Intelligent Functional Safety Engineering. Aegis is specifically designed to support complex functional safety tasks within the automotive sector. It is tailored to perform Hazard Analysis and Risk Assessment (HARA), document Functional Safety Requirements (FSR), and plan test cases for Automatic Emergency Braking (AEB) systems. The most advanced version, Aegis-Max, leverages Retrieval-Augmented Generation (RAG) and reflective mechanisms to enhance its capability in managing complex, knowledge-intensive tasks. Additionally, targeted prompt refinement by professional functional safety practitioners can significantly optimize Aegis’s performance in the functional safety domain. This paper demonstrates the potential of Aegis to improve the efficiency and effectiveness of functional safety processes in automotive engineering.

pdf bib abs
Efficient Answer Retrieval System (EARS): Combining Local DB Search and Web Search for Generative QA
Nikita Krayko | Ivan Sidorov | Fedor Laputin | Daria Galimzianova | Vasily Konovalov

In this work, we propose an efficient answer retrieval system **EARS**: a production-ready, factual question answering (QA) system that combines local knowledge base search with generative, context-based QA. To assess the quality of the generated content, we devise comprehensive metrics for both manual and automatic evaluation of the answers to questions. A distinctive feature of our system is the Ranker component, which ranks answer candidates based on their relevance. This feature enhances the effectiveness of local knowledge base retrieval by 23%. Another crucial aspect of our system is the LLM, which utilizes contextual information from a web search API to generate responses. This results in substantial 92.8% boost in the usefulness of voice-based responses. **EARS** is language-agnostic and can be applied to any data domain.

GraphQL is a powerful query language for APIs that allows clients to fetch precise data efficiently and flexibly, querying multiple resources with a single request. However, crafting complex GraphQL query operations can be challenging. Large Language Models (LLMs) offer an alternative by generating GraphQL queries from natural language, but they struggle due to limited exposure to publicly available GraphQL schemas, often resulting in invalid or suboptimal queries. Furthermore, no benchmark test data suite is available to reliably evaluate the performance of contemporary LLMs.To address this, we present a large-scale, cross-domain Text-to-GraphQL query operation dataset. The dataset includes 10,940 training triples spanning 185 cross-source data stores and 957 test triples over 14 data stores. Each triple consists of a GraphQL schema, GraphQL query operation, and corresponding natural language query. The dataset has been predominantly manually created, with natural language paraphrasing, and carefully validated, requiring approximately 1200 person-hours. In our evaluation, we tested 10 state-of-the-art LLMs using our test dataset. The best-performing model achieved an accuracy of only around 50% with one in-context few-shot example, underscoring the necessity for custom fine-tuning. To support further research and benchmarking, we are releasing the training and test datasets under the MIT License. The dataset is available at https://github.com/stepzen-dev/NL2GQL.

pdf bib abs
Mixture of Diverse Size Experts
Manxi Sun | Wei Liu | Jian Luan | Pengzhi Gao | Bin Wang

The Sparsely-Activated Mixture-of-Experts (MoE) architecture has gained popularity for scaling large language models (LLMs) due to the sub-linearly increasing computational costs. Despite its success, most of the current structure designs face the challenge that the experts share the same size such that tokens have no chance to choose the experts with the most appropriate size to generate the next token. To migrate this defect, we propose Mixture of Diverse Size Experts (MoDSE), a new MoE architecture with designed layers where experts have different sizes. Analysis on difficult token generation tasks shows that experts with different sizes give better predictions, and the routing path of the experts tends to be stable after a period of training. The diversity of experts’ size will lead to load unbalancing. To tackle this limitation, we introduce an expert-pair allocation strategy to distribute the workload evenly across the GPUs. Comprehensive evaluations across multiple benchmarks demonstrate the effectiveness of MoDSE, surpassing existing MoEs by adaptively assigning the parameter budget to experts while maintaining the same total parameter size and number of experts.

The risk of harmful contents generated by large language models (LLMs) becomes a critical concern. This paper systematically evaluates and enhances LLMs’ capability to perform course-correction, , the model can steer away from generating harmful content autonomously. First, we introduce the C²-Eval benchmark for quantitative assessment and analyze 10 popular LLMs, revealing varying proficiency of current safety-tuned LLMs in course-correction.To improve, we propose fine-tuning LLMs with preference learning, emphasizing the preference for timely course-correction. Using an automated pipeline, we create C²-Syn, a synthetic C²-Syn with 750K pairwise preferences, to teach models the concept of timely course-correction through data-driven learning.Experiments on Llama2-Chat 7B and Qwen2 7B show that our method effectively enhances course-correction skills without affecting general performance. Additionally, it effectively improves LLMs’ safety, particularly in resisting jailbreak attacks.

Pre-trained language models have become an integral component of question-answering systems, achieving remarkable performance. However, for practical deployment, it is crucial to perform knowledge distillation to maintain high performance while operating under computational constraints. In this paper, we address a key question: given the importance of unsupervised distillation for student model performance, how can knowledge from multiple teacher models be effectively ensemble during this stage without the guidance of labels? We propose a novel algorithm, GOVERN, to tackle this issue. GOVERN has demonstrated significant improvements in both offline and online experiments, enabling the student model to achieve results comparable to that of teacher ensembles. Our experiments show that GOVERN remarkably requires a mere 1% of the ensemble method’s inference budget to achieve 99.5% of performance. The proposed algorithm has been successfully deployed in a real-world commercial question-answering system, demonstrating its real-world applicability.

While image-text pre-trained models, such as CLIP, have demonstrated impressive capabilities in learning robust text and image representations, a critical area for substantial improvement remains—precise color understanding. In this paper, we address this limitation by introducing PRISM, a simple yet highly effective method that extends CLIP’s capability to grasp the nuances of precise colors. PRISM seamlessly adapts to both recognized HTML colors and out-of-vocabulary RGB inputs through the utilization of our curated dataset of 100 image-text pairs, which can be effortlessly repurposed for fine-tuning with any desired color. Importantly, PRISM achieves these enhancements without compromising CLIP’s performance on established benchmarks. Furthermore, we introduce a novel evaluation framework, ColorLens, featuring both seen and unseen test sets that can be readily repurposed to assess a model’s precision in understanding precise colors. Our comprehensive evaluation and results demonstrate significant improvements over baseline models.