Large language models (LLMs), especially when instruction-tuned for chat, have become part of our daily lives, freeing people from the process of searching, extracting, and integrating information from multiple sources by offering a straightforward answer to a variety of questions in a single place. Unfortunately, in many cases, LLM responses are factually incorrect, which limits their applicability in real-world scenarios. As a result, research on evaluating and improving the factuality of LLMs has attracted a lot of research attention recently. In this survey, we critically analyze existing work with the aim to identify the major challenges and their associated causes, pointing out to potential solutions for improving the factuality of LLMs, and analyzing the obstacles to automated factuality evaluation for open-ended text generation. We further offer an outlook on where future research should go.

The increased use of large language models (LLMs) across a variety of real-world applications calls for automatic tools to check the factual accuracy of their outputs, as LLMs often hallucinate. This is difficult as it requires assessing the factuality of free-form open-domain responses. While there has been a lot of research on this topic, different papers use different evaluation benchmarks and measures,which makes them hard to compare and hampers future progress. To mitigate these issues, we developed OpenFactCheck, a unified framework, with three modules: (i) RESPONSEEVAL, which allows users to easily customize an automatic fact-checking system and to assess the factuality of all claims in an input document using that system, (ii) LLMEVAL, which assesses the overall factuality of an LLM, and (iii) CHECKEREVAL, a module to evaluate automatic fact-checking systems. OpenFactCheck is open-sourced (https://github.com/mbzuai-nlp/openfactcheck) and publicly released as a Python library (https://pypi.org/project/openfactcheck/) and also as a web service (http://app.openfactcheck.com). A video describing the system is available at https://youtu.be/-i9VKL0HleI.

The ease of access to large language models (LLMs) has enabled a widespread of machine-generated texts, and now it is often hard to tell whether a piece of text was human-written or machine-generated. This raises concerns about potential misuse, particularly within educational and academic domains. Thus, it is important to develop practical systems that can automate the process. Here, we present one such system, LLM-DetectAIve, designed for fine-grained detection. Unlike most previous work on machine-generated text detection, which focused on binary classification, LLM-DetectAIve supports four categories: (i) human-written, (ii) machine-generated, (iii) machine-written, then machine-humanized, and (iv) human-written, then machine-polished. Category (iii) aims to detect attempts to obfuscate the fact that a text was machine-generated, while category (iv) looks for cases where the LLM was used to polish a human-written text, which is typically acceptable in academic writing, but not in education. Our experiments show that LLM-DetectAIve can effectively identify the above four categories, which makes it a potentially useful tool in education, academia, and other domains.LLM-DetectAIve is publicly accessible at https://github.com/mbzuai-nlp/LLM-DetectAIve. The video describing our system is available at https://youtu.be/E8eT_bE7k8c.

With the rapid evolution of large language models (LLMs), new and hard-to-predict harmful capabilities are emerging. This requires developers to identify potential risks through the evaluation of “dangerous capabilities” in order to responsibly deploy LLMs. Here we aim to facilitate this process. In particular, we collect an open-source dataset to evaluate the safeguards in LLMs, to facilitate the deployment of safer open-source LLMs at a low cost. Our dataset is curated and filtered to consist only of instructions that responsible language models should not follow. We assess the responses of six popular LLMs to these instructions, and we find that simple BERT-style classifiers can achieve results that are comparable to GPT-4 on automatic safety evaluation. Our data and code are available at https://github.com/Libr-AI/do-not-answer

Recent years, have seen the rise of large language models (LLMs), where practitioners use task-specific prompts; this was shown to be effective for a variety of tasks. However, when applied to semantic textual similarity (STS) and natural language inference (NLI), the effectiveness of LLMs turns out to be limited by low-resource domain accuracy, model overconfidence, and difficulty to capture the disagreements between human judgements. With this in mind, here we try to rethink STS and NLI in the era of LLMs. We first evaluate the performance of STS and NLI in the clinical/biomedical domain, and then we assess LLMs’ predictive confidence and their capability of capturing collective human opinions. We find that these old problems are still to be properly addressed in the era of LLMs.

Many studies have demonstrated that large language models (LLMs) can produce harmful responses, exposing users to unexpected risks. Previous studies have proposed comprehensive taxonomies of LLM risks, as well as corresponding prompts that can be used to examine LLM safety. However, the focus has been almost exclusively on English. We aim to broaden LLM safety research by introducing a dataset for the safety evaluation of Chinese LLMs, and extending it to better identify false negative and false positive examples in terms of risky prompt rejections. We further present a set of fine-grained safety assessment criteria for each risk type, facilitating both manual annotation and automatic evaluation in terms of LLM response harmfulness. Our experiments over five LLMs show that region-specific risks are the prevalent risk type. Warning: this paper contains example data that may be offensive, harmful, or biased. Our data is available at https://github.com/Libr-AI/do-not-answer.

Recent advancements in multimodal large language models (MLLMs) have made significant progress in integrating information across various modalities, yet real-world applications in educational and scientific domains remain challenging. This paper introduces the Multimodal Scientific ASR (MS-ASR) task, which focuses on transcribing scientific conference videos by leveraging visual information from slides to enhance the accuracy of technical terminologies. Realized that traditional metrics like WER fall short in assessing performance accurately, prompting the proposal of severity-aware WER (SWER) that considers the content type and severity of ASR errors. We propose the Scientific Vision Augmented ASR (SciVASR) framework as a baseline method, enabling MLLMs to improve transcript quality through post-editing. Evaluations of state-of-the-art MLLMs, including GPT-4o, show a 45% improvement over speech-only baselines, highlighting the importance of multimodal information integration.

The increased use of large language models (LLMs) across a variety of real-world applications calls for mechanisms to verify the factual accuracy of their outputs. In this work, we present Factcheck-Bench, a holistic end-to-end framework for annotating and evaluating the factuality of LLM-generated responses, which encompasses a multi-stage annotation scheme designed to yield detailed labels for fact-checking and correcting not just the final prediction, but also the intermediate steps that a fact-checking system might need to take. Based on this framework, we construct an open-domain factuality benchmark in three-levels of granularity: claim, sentence, and document. We further propose a system, Factcheck-GPT, which follows our framework, and we show that it outperforms several popular LLM fact-checkers. We make our annotation tool, annotated data, benchmark, and code available at https://github.com/yuxiaw/Factcheck-GPT.

Empathy plays a pivotal role in fostering prosocial behavior, often triggered by the sharing of personal experiences through narratives. However, modeling empathy using NLP approaches remains challenging due to its deep interconnection with human interaction dynamics. Previous approaches, which involve fine-tuning language models (LMs) on human-annotated empathic datasets, have had limited success. In our pursuit of improving empathy understanding in LMs, we propose several strategies, including contrastive learning with masked LMs and supervised fine-tuning with large language models. While these methods show improvements over previous methods, the overall results remain unsatisfactory. To better understand this trend, we performed an analysis which reveals a low agreement among annotators. This lack of consensus hinders training and highlights the subjective nature of the task. We also explore the cultural impact on annotations. To study this, we meticulously collected story pairs in Urdu language and find that subjectivity in interpreting empathy among annotators appears to be independent of cultural background. Our systematic exploration of LMs’ understanding of empathy reveals substantial opportunities for further investigation in both task formulation and modeling.

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks in various domains. Despite their impressive performance, they can be unreliable due to factual errors in their generations. Assessing their confidence and calibrating them across different tasks can help mitigate risks and enable LLMs to produce better generations. There has been a lot of recent research aiming to address this, but there has been no comprehensive overview to organize it and to outline the main lessons learned. The present survey aims to bridge this gap. In particular, we outline the challenges and we summarize recent technical advancements for LLM confidence estimation and calibration. We further discuss their applications and suggest promising directions for future work.

Large language models (LLMs) have demonstrated remarkable capability to generate fluent responses to a wide variety of user queries. However, this has also raised concerns about the potential misuse of such texts in journalism, education, and academia. In this study, we strive to create automated systems that can detect machine-generated texts and pinpoint potential misuse. We first introduce a large-scale benchmark M4, which is a multi-generator, multi-domain, and multi-lingual corpus for machine-generated text detection. Through an extensive empirical study of this dataset, we show that it is challenging for detectors to generalize well on instances from unseen domains or LLMs. In such cases, detectors tend to misclassify machine-generated text as human-written. These results show that the problem is far from solved and that there is a lot of room for improvement. We believe that our dataset will enable future research towards more robust approaches to this pressing societal problem. The dataset is available at https://github.com/mbzuai-nlp/M4

The advent of Large Language Models (LLMs) has brought an unprecedented surge in machine-generated text (MGT) across diverse channels. This raises legitimate concerns about its potential misuse and societal implications. The need to identify and differentiate such content from genuine human-generated text is critical in combating disinformation, preserving the integrity of education and scientific fields, and maintaining trust in communication. In this work, we address this problem by introducing a new benchmark based on a multilingual, multi-domain and multi-generator corpus of MGTs — M4GT-Bench. The benchmark is compiled of three tasks: (1) mono-lingual and multi-lingual binary MGT detection; (2) multi-way detection where one need to identify, which particular model generated the text; and (3) mixed human-machine text detection, where a word boundary delimiting MGT from human-written content should be determined. On the developed benchmark, we have tested several MGT detection baselines and also conducted an evaluation of human performance. We see that obtaining good performance in MGT detection usually requires an access to the training data from the same domain and generators. The benchmark is available at https://github.com/mbzuai-nlp/M4GT-Bench.

Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore the effects of instruction tuning on different combinations of datasets on LLM performance, and find that certain instruction types are more advantageous for specific applications but can negatively impact other areas. This work provides insights into instruction mixtures, laying the foundations for future research.

We present the results and the main findings of SemEval-2024 Task 8: Multigenerator, Multidomain, and Multilingual Machine-Generated Text Detection. The task featured three subtasks. Subtask A is a binary classification task determining whether a text is written by a human or generated by a machine. This subtask has two tracks: a monolingual track focused solely on English texts and a multilingual track. Subtask B is to detect the exact source of a text, discerning whether it is written by a human or generated by a specific LLM. Subtask C aims to identify the changing point within a text, at which the authorship transitions from human to machine. The task attracted a large number of participants: subtask A monolingual (126), subtask A multilingual (59), subtask B (70), and subtask C (30). In this paper, we present the task, analyze the results, and discuss the system submissions and the methods they used. For all subtasks, the best systems used LLMs.

Despite the subjective nature of semantic textual similarity (STS) and pervasive disagreements in STS annotation, existing benchmarks have used averaged human ratings as gold standard. Averaging masks the true distribution of human opinions on examples of low agreement, and prevents models from capturing the semantic vagueness that the individual ratings represent. In this work, we introduce USTS, the first Uncertainty-aware STS dataset with ∼15,000 Chinese sentence pairs and 150,000 labels, to study collective human opinions in STS. Analysis reveals that neither a scalar nor a single Gaussian fits a set of observed judgments adequately. We further show that current STS models cannot capture the variance caused by human disagreement on individual instances, but rather reflect the predictive confidence over the aggregate dataset.

Natural Language Inference (NLI) datasets contain examples with highly ambiguous labels due to its subjectivity. Several recent efforts have been made to acknowledge and embrace the existence of ambiguity, and explore how to capture the human disagreement distribution. In contrast with directly learning from gold ambiguity labels, relying on special resource, we argue that the model has naturally captured the human ambiguity distribution as long as it’s calibrated, i.e. the predictive probability can reflect the true correctness likelihood. Our experiments show that when model is well-calibrated, either by label smoothing or temperature scaling, it can obtain competitive performance as prior work, on both divergence scores between predictive probability and the true human opinion distribution, and the accuracy. This reveals the overhead of collecting gold ambiguity labels can be cut, by broadly solving how to calibrate the NLI network.

Autoregressive (AR) and Non-autoregressive (NAR) models have their own superiority on the performance and latency, combining them into one model may take advantage of both. Current combination frameworks focus more on the integration of multiple decoding paradigms with a unified generative model, e.g. Masked Language Model. However, the generalization can be harmful on the performance due to the gap between training objective and inference. In this paper, we aim to close the gap by preserving the original objective of AR and NAR under a unified framework. Specifically, we propose the Directional Transformer (Diformer) by jointly modelling AR and NAR into three generation directions (left-to-right, right-to-left and straight) with a newly introduced direction variable, which works by controlling the prediction of each token to have specific dependencies under that direction. The unification achieved by direction successfully preserves the original dependency assumption used in AR and NAR, retaining both generalization and performance. Experiments on 4 WMT benchmarks demonstrate that Diformer outperforms current united-modelling works with more than 1.5 BLEU points for both AR and NAR decoding, and is also competitive to the state-of-the-art independent AR and NAR models.

State-of-the-art classification and regression models are often not well calibrated, and cannot reliably provide uncertainty estimates, limiting their utility in safety-critical applications such as clinical decision-making. While recent work has focused on calibration of classifiers, there is almost no work in NLP on calibration in a regression setting. In this paper, we quantify the calibration of pre- trained language models for text regression, both intrinsically and extrinsically. We further apply uncertainty estimates to augment training data in low-resource domains. Our experiments on three regression tasks in both self-training and active-learning settings show that uncertainty estimation can be used to increase overall performance and enhance model generalization.

This paper describes the HW-TSC’s designation of the Offline Speech Translation System submitted for IWSLT 2022 Evaluation. We explored both cascade and end-to-end system on three language tracks (en-de, en-zh and en-ja), and we chose the cascade one as our primary submission. For the automatic speech recognition (ASR) model of cascade system, there are three ASR models including Conformer, S2T-Transformer and U2 trained on the mixture of five datasets. During inference, transcripts are generated with the help of domain controlled generation strategy. Context-aware reranking and ensemble based anti-interference strategy are proposed to produce better ASR outputs. For machine translation part, we pretrained three translation models on WMT21 dataset and fine-tuned them on in-domain corpora. Our cascade system shows competitive performance than the known offline systems in the industry and academia.

This paper presents our work in the participation of IWSLT 2022 simultaneous speech translation evaluation. For the track of text-to-text (T2T), we participate in three language pairs and build wait-k based simultaneous MT (SimulMT) model for the task. The model was pretrained on WMT21 news corpora, and was further improved with in-domain fine-tuning and self-training. For the speech-to-text (S2T) track, we designed both cascade and end-to-end form in three language pairs. The cascade system is composed of a chunking-based streaming ASR model and the SimulMT model used in the T2T track. The end-to-end system is a simultaneous speech translation (SimulST) model based on wait-k strategy, which is directly trained on a synthetic corpus produced by translating all texts of ASR corpora into specific target language with an offline MT model. It also contains a heuristic sentence breaking strategy, preventing it from finishing the translation before the the end of the speech. We evaluate our systems on the MUST-C tst-COMMON dataset and show that the end-to-end system is competitive to the cascade one. Meanwhile, we also demonstrate that the SimulMT model can be efficiently optimized by these approaches, resulting in the improvements of 1-2 BLEU points.

The paper presents the HW-TSC’s pipeline and results of Offline Speech to Speech Translation for IWSLT 2022. We design a cascade system consisted of an ASR model, machine translation model and TTS model to convert the speech from one language into another language(en-de). For the ASR part, we find that better performance can be obtained by ensembling multiple heterogeneous ASR models and performing reranking on beam candidates. And we find that the combination of context-aware reranking strategy and MT model fine-tuned on the in-domain dataset is helpful to improve the performance. Because it can mitigate the problem that the inconsistency in transcripts caused by the lack of context. Finally, we use VITS model provided officially to reproduce audio files from the translation hypothesis.

Training with noisy labelled data is known to be detrimental to model performance, especially for high-capacity neural network models in low-resource domains. Our experiments suggest that standard regularisation strategies, such as weight decay and dropout, are ineffective in the face of noisy labels. We propose a simple noisy label detection method that prevents error propagation from the input layer. The approach is based on the observation that the projection of noisy labels is learned through memorisation at advanced stages of learning, and that the Pearson correlation is sensitive to outliers. Extensive experiments over real-world human-disagreement annotations as well as randomly-corrupted and data-augmented labels, across various tasks and domains, demonstrate that our method is effective, regularising noisy labels and improving generalisation performance.

Length prediction is a special task in a series of NAT models where target length has to be determined before generation. However, the performance of length prediction and its influence on translation quality has seldom been discussed. In this paper, we present comprehensive analyses on length prediction task of NAT, aiming to find the factors that influence performance, as well as how it associates with translation quality. We mainly perform experiments based on Conditional Masked Language Model (CMLM) (Ghazvininejad et al., 2019), a representative NAT model, and evaluate it on two language pairs, En-De and En-Ro. We draw two conclusions: 1) The performance of length prediction is mainly influenced by properties of language pairs such as alignment pattern, word order or intrinsic length ratio, and is also affected by the usage of knowledge distilled data. 2) There is a positive correlation between the performance of the length prediction and the BLEU score.

Mask-predict CMLM (Ghazvininejad et al.,2019) has achieved stunning performance among non-autoregressive NMT models, but we find that the mechanism of predicting all of the target words only depending on the hidden state of [MASK] is not effective and efficient in initial iterations of refinement, resulting in ungrammatical repetitions and slow convergence. In this work, we mitigate this problem by combining copied source with embeddings of [MASK] in decoder. Notably. it’s not a straightforward copying that is shown to be useless, but a novel heuristic hybrid strategy — fence-mask. Experimental results show that it gains consistent boosts on both WMT14 En<->De and WMT16 En<->Ro corpus by 0.5 BLEU on average, and 1 BLEU for less-informative short sentences. This reveals that incorporating additional information by proper strategies is beneficial to improve CMLM, particularly translation quality of short texts and speeding up early-stage convergence.

This paper presents our work in WMT 2021 Quality Estimation (QE) Shared Task. We participated in all of the three sub-tasks, including Sentence-Level Direct Assessment (DA) task, Word and Sentence-Level Post-editing Effort task and Critical Error Detection task, in all language pairs. Our systems employ the framework of Predictor-Estimator, concretely with a pre-trained XLM-Roberta as Predictor and task-specific classifier or regressor as Estimator. For all tasks, we improve our systems by incorporating post-edit sentence or additional high-quality translation sentence in the way of multitask learning or encoding it with predictors directly. Moreover, in zero-shot setting, our data augmentation strategy based on Monte-Carlo Dropout brings up significant improvement on DA sub-task. Notably, our submissions achieve remarkable results over all tasks.

Domain pretraining followed by task fine-tuning has become the standard paradigm for NLP tasks, but requires in-domain labelled data for task fine-tuning. To overcome this, we propose to utilise domain unlabelled data by assigning pseudo labels from a general model. We evaluate the approach on two clinical STS datasets, and achieve r= 0.80 on N2C2-STS. Further investigation reveals that if the data distribution of unlabelled sentence pairs is closer to the test data, we can obtain better performance. By leveraging a large general-purpose STS dataset and small-scale in-domain training data, we obtain further improvements to r= 0.90, a new SOTA.

In this paper, we apply pre-trained language models to the Semantic Textual Similarity (STS) task, with a specific focus on the clinical domain. In low-resource setting of clinical STS, these large models tend to be impractical and prone to overfitting. Building on BERT, we study the impact of a number of model design choices, namely different fine-tuning and pooling strategies. We observe that the impact of domain-specific fine-tuning on clinical STS is much less than that in the general domain, likely due to the concept richness of the domain. Based on this, we propose two data augmentation techniques. Experimental results on N2C2-STS 1 demonstrate substantial improvements, validating the utility of the proposed methods.