Scientific reasoning poses an excessive challenge for even the most advanced Large Language Models (LLMs). To make this task more practical and solvable for LLMs, we introduce a new task setting named tool-augmented scientific reasoning. This setting supplements LLMs with scalable toolsets, and shifts the focus from pursuing an omniscient problem solver to a proficient tool-user. To facilitate the research of such setting, we construct a tool-augmented training corpus named MathFunc which encompasses over 30,000 samples and roughly 6,000 tools. Building on MathFunc, we develop SciAgent to retrieve, understand and, if necessary, use tools for scientific problem solving. Additionally, we craft a benchmark, SciToolBench, spanning five scientific domains to evaluate LLMs’ abilities with tool assistance. Extensive experiments on SciToolBench confirm the effectiveness of SciAgent. Notably, SciAgent-Llama3-8B surpasses other LLMs with the comparable size by more than 8.0% in absolute accuracy. Furthermore, SciAgent-DeepMath-7B shows much superior performance than ChatGPT.

Although achieving great success, Large Language Models (LLMs) usually suffer from unreliable hallucinations. Although language attribution can be a potential solution, there are no suitable benchmarks and evaluation metrics to attribute LLMs to structured knowledge. In this paper, we define a new task of Knowledge-aware Language Model Attribution (KaLMA) that improves upon three core concerns with conventional attributed LMs. First, we extend attribution source from unstructured texts to Knowledge Graph (KG), whose rich structures benefit both the attribution performance and working scenarios. Second, we propose a new “Conscious Incompetence” setting considering the incomplete knowledge repository, where the model identifies the need for supporting knowledge beyond the provided KG. Third, we propose a comprehensive automatic evaluation metric encompassing text quality, citation quality, and text citation alignment. To implement the above innovations, we build a dataset in biography domain BioKaLMA via evolutionary question generation strategy, to control the question complexity and necessary knowledge to the answer. For evaluation, we develop a baseline solution and demonstrate the room for improvement in LLMs’ citation generation, emphasizing the importance of incorporating the “Conscious Incompetence” setting, and the critical role of retrieval accuracy.

Conversational recommender systems (CRS) aim to recommend relevant items to users by eliciting user preference through natural language conversation. Prior work often utilizes external knowledge graphs for items’ semantic information, a language model for dialogue generation, and a recommendation module for ranking relevant items. This combination of multiple components suffers from a cumber-some training process, and leads to semantic misalignment issues between dialogue generation and item recommendation. In this paper, we represent items in natural language and formulate CRS as a natural language processing task. Accordingly, we leverage the power of pre-trained language models to encode items, understand user intent via conversation, perform item recommendation through semantic matching, and generate dialogues. As a unified model, our PECRS (Parameter-Efficient CRS), can be optimized in a single stage, without relying on non-textual metadata such as a knowledge graph. Experiments on two benchmark CRS datasets, ReDial and INSPIRED, demonstrate the effectiveness of PECRS on recommendation and conversation. Our code is available at: https://github.com/Ravoxsg/efficient_unified_crs.

Large language models (LLMs) excel in abstractive summarization tasks, delivering fluent and pertinent summaries. Recent advancements have extended their capabilities to handle long-input contexts, exceeding 100k tokens. However, in question answering, language models exhibit uneven utilization of their input context. They tend to favor the initial and final segments, resulting in a U-shaped performance pattern concerning where the answer is located within the input. This bias raises concerns, particularly in summarization where crucial content may be dispersed throughout the source document(s). Besides, in summarization, mapping facts from the source to the summary is not trivial as salient content is usually re-phrased. In this paper, we conduct the first comprehensive study on context utilization and position bias in summarization. Our analysis encompasses 6 LLMs, 10 datasets, and 5 evaluation metrics. We introduce a new evaluation benchmark called MiddleSum on the which we benchmark two alternative inference methods to alleviate position bias: hierarchical summarization and incremental summarization. Our code and data can be found here: https://github.com/ntunlp/MiddleSum.

A crucial aspect in abusive language on social media platforms (toxicity, hate speech, harmful stereotypes, etc.) is its inherent contextual nature. In this paper, we focus on the role of conversational context in abusive language detection, one of the most “direct” forms of context in this domain, as given by the conversation threads (e.g., directly preceding message, original post). The incorporation of surrounding messages has proven vital for the accurate human annotation of harmful content. However, many prior works have either ignored this aspect, collecting and processing messages in isolation, or have obtained inconsistent results when attempting to embed such contextual information into traditional classification methods. The reasons behind these findings have not yet been properly addressed. To this end, we propose an analysis of the impact of conversational context in abusive language detection, through: (1) an analysis of prior works and the limitations of the most common concatenation-based approach, which we attempt to address with two alternative architectures; (2) an evaluation of these methods on existing datasets in English, and a new dataset of French tweets annotated for hate speech and stereotypes; and (3) a qualitative analysis showcasing the necessity for context-awareness in ALD, but also its difficulties.

Given a text query, the task of Natural Language Video Localization (NLVL) is to localize a temporal moment in an untrimmed video that semantically matches the query. In this paper, we adopt a proposal-based solution that generates proposals (i.e. candidate moments) and then select the best matching proposal. On top of modeling the cross-modal interaction between candidate moments and the query, our proposed Moment Sampling DETR (MS-DETR) enables efficient moment-moment relation modeling. The core idea is to sample a subset of moments guided by the learnable templates with an adopted DETR framework. To achieve this, we design a multi-scale visual-linguistic encoder, and an anchor-guided moment decoder paired with a set of learnable templates. Experimental results on three public datasets demonstrate the superior performance of MS-DETR.

Training grounded response generation models often requires a large collection of grounded dialogues. However, it is costly to build such dialogues. In this paper, we present a synthetic data generation framework (SynDG) for grounded dialogues. The generation process utilizes large pre-trained language models and freely available knowledge data (e.g., Wikipedia pages, persona profiles, etc.). The key idea of designing SynDG is to consider dialogue flow and coherence in the generation process. Specifically, given knowledge data, we first heuristically determine a dialogue flow, which is a series of knowledge pieces. Then, we employ T5 to incrementally turn the dialogue flow into a dialogue. To ensure coherence of both the dialogue flow and the synthetic dialogue, we design a two-level filtering strategy, at the flow-level and the utterance-level respectively. Experiments on two public benchmarks show that the synthetic grounded dialogue data produced by our framework is able to significantly boost model performance in both full training data and low-resource scenarios.

Few-shot event detection (ED) has been widely studied, while this brings noticeable discrepancies, e.g., various motivations, tasks, and experimental settings, that hinder the understanding of models for future progress. This paper presents a thorough empirical study, a unified view of ED models, and a better unified baseline. For fair evaluation, we compare 12 representative methods on three datasets, which are roughly grouped into prompt-based and prototype-based models for detailed analysis. Experiments consistently demonstrate that prompt-based methods, including ChatGPT, still significantly trail prototype-based methods in terms of overall performance. To investigate their superior performance, we break down their design elements along several dimensions and build a unified framework on prototype-based methods. Under such unified view, each prototype-method can be viewed a combination of different modules from these design elements. We further combine all advantageous modules and propose a simple yet effective baseline, which outperforms existing methods by a large margin (e.g., 2.7% F1 gains under low-resource setting).

Negation in natural language may affect many NLP applications, e.g., information extraction and sentiment analysis. The key sub-task of negation detection is negation scope resolution which aims to extract the portion of a sentence that is being negated by a negation cue (e.g., keyword “not” and never”) in the sentence. Due to the long spans, existing methods tend to make wrong predictions around the scope boundaries. In this paper, we propose a simple yet effective model named R-BSL which engages the Boundary Shift Loss to refine the predicted boundary. On multiple benchmark datasets, we show that the extremely simple R-BSL achieves best results.

Goal-oriented Script Generation is a new task of generating a list of steps that can fulfill the given goal. In this paper, we propose to extend the task from the perspective of cognitive theory. Instead of a simple flat structure, the steps are typically organized hierarchically — Human often decompose a complex task into subgoals, where each subgoal can be further decomposed into steps. To establish the benchmark, we contribute a new dataset, propose several baseline methods, and set up evaluation metrics. Both automatic and human evaluation verify the high-quality of dataset, as well as the effectiveness of incorporating subgoals into hierarchical script generation. Furthermore, We also design and evaluate the model to discover subgoal, and find that it is a bit more difficult to decompose the goals than summarizing from segmented steps.

Knowledge Graph Completion (KGC) often requires both KG structural and textual information to be effective. Pre-trained Language Models (PLMs) have been used to learn the textual information, usually under the fine-tune paradigm for the KGC task. However, the fine-tuned PLMs often overwhelmingly focus on the textual information and overlook structural knowledge. To tackle this issue, this paper proposes CSProm-KG (Conditional Soft Prompts for KGC) which maintains a balance between structural information and textual knowledge. CSProm-KG only tunes the parameters of Conditional Soft Prompts that are generated by the entities and relations representations. We verify the effectiveness of CSProm-KG on three popular static KGC benchmarks WN18RR, FB15K-237 and Wikidata5M, and two temporal KGC benchmarks ICEWS14 and ICEWS05-15. CSProm-KG outperforms competitive baseline models and sets new state-of-the-art on these benchmarks. We conduct further analysis to show (i) the effectiveness of our proposed components, (ii) the efficiency of CSProm-KG, and (iii) the flexibility of CSProm-KG.

Large Language Models (LLMs) have made remarkable strides in various tasks. Whether LLMs are competitive few-shot solvers for information extraction (IE) tasks, however, remains an open problem. In this work, we aim to provide a thorough answer to this question. Through extensive experiments on nine datasets across four IE tasks, we demonstrate that current advanced LLMs consistently exhibit inferior performance, higher latency, and increased budget requirements compared to fine-tuned SLMs under most settings. Therefore, we conclude that LLMs are not effective few-shot information extractors in general. Nonetheless, we illustrate that with appropriate prompting strategies, LLMs can effectively complement SLMs and tackle challenging samples that SLMs struggle with. And moreover, we propose an adaptive filter-then-rerank paradigm to combine the strengths of LLMs and SLMs. In this paradigm, SLMs serve as filters and LLMs serve as rerankers. By prompting LLMs to rerank a small portion of difficult samples identified by SLMs, our preliminary system consistently achieves promising improvements (2.4% F1-gain on average) on various IE tasks, with an acceptable time and cost investment.

Speculation detection is an important NLP task to identify text factuality. However, the extracted speculative information (e.g., speculative polarity, cue, and scope) lacks structure and poses challenges for direct utilization in downstream tasks. Open Information Extraction (OIE), on the other hand, extracts structured tuples as facts, without examining the certainty of these tuples. Bridging this gap between speculation detection and information extraction becomes imperative to generate structured speculative information and trustworthy relational tuples. Existing studies on speculation detection are defined at sentence level; but even if a sentence is determined to be speculative, not all factual tuples extracted from it are speculative. In this paper, we propose to study speculations in OIE tuples and determine whether a tuple is speculative. We formally define the research problem of tuple-level speculation detection. We then conduct detailed analysis on the LSOIE dataset which provides labels for speculative tuples. Lastly, we propose a baseline model SpecTup for this new research task.

Open Information Extraction (OIE) aims to extract relational tuples from open-domain sentences. Existing OIE systems split a sentence into tokens and recognize token spans as tuple relations and arguments. We instead propose Sentence as Chunk sequence (SaC) and recognize chunk spans as tuple relations and arguments. We argue that SaC has better properties for OIE than sentence as token sequence, and evaluate four choices of chunks (i.e., CoNLL chunks, OIA simple phrases, noun phrases, and spans from SpanOIE). Also, we propose a simple end-to-end BERT-based model, Chunk-OIE, for sentence chunking and tuple extraction on top of SaC. Chunk-OIE achieves state-of-the-art results on multiple OIE datasets, showing that SaC benefits the OIE task.

Events are fundamental building blocks of real-world happenings. In this paper, we present a large-scale, multi-modal event knowledge graph named MMEKG. MMEKG unifies different modalities of knowledge via events, which complement and disambiguate each other. Specifically, MMEKG incorporates (i) over 990 thousand concept events with 644 relation types to cover most types of happenings, and (ii) over 863 million instance events connected through 934 million relations, which provide rich contextual information in texts and/or images. To collect billion-scale instance events and relations among them, we additionally develop an efficient yet effective pipeline for textual/visual knowledge extraction system. We also develop an induction strategy to create million-scale concept events and a schema organizing all events and relations in MMEKG. To this end, we also provide a pipeline enabling our system to seamlessly parse texts/images to event graphs and to retrieve multi-modal knowledge at both concept- and instance-levels.

Open Information Extraction (OpenIE) aims to extract relational tuples from open-domain sentences. Traditional rule-based or statistical models were developed based on syntactic structure of sentence, identified by syntactic parsers. However, previous neural OpenIE models under-explored the useful syntactic information. In this paper, we model both constituency and dependency trees into word-level graphs, and enable neural OpenIE to learn from the syntactic structures. To better fuse heterogeneous information from the two graphs, we adopt multi-view learning to capture multiple relationships from them. Finally, the finetuned constituency and dependency representations are aggregated with sentential semantic representations for tuple generation. Experiments show that both constituency and dependency information, and the multi-view learning are effective.

Sarcasm employs ambivalence, where one says something positive but actually means negative, and vice versa. The essence of sarcasm, which is also a sufficient and necessary condition, is the conflict between literal and implied sentiments expressed in one sentence. However, it is difficult to recognize such sentiment conflict because the sentiments are mixed or even implicit. As a result, the recognition of sophisticated and obscure sentiment brings in a great challenge to sarcasm detection. In this paper, we propose a Dual-Channel Framework by modeling both literal and implied sentiments separately. Based on this dual-channel framework, we design the Dual-Channel Network (DC-Net) to recognize sentiment conflict. Experiments on political debates (i.e. IAC-V1 and IAC-V2) and Twitter datasets show that our proposed DC-Net achieves state-of-the-art performance on sarcasm recognition. Our code is released to support research.

Comparative opinion is a common linguistic phenomenon. The opinion is expressed by comparing multiple targets on a shared aspect, e.g., “camera A is better than camera B in picture quality”. Among the various subtasks in opinion mining, comparative opinion classification is relatively less studied. Current solutions use rules or classifiers to identify opinions, i.e., better, worse, or same, through feature engineering. Because the features are directly derived from the input sentence, these solutions are sensitive to the order of the targets mentioned in the sentence. For example, “camera A is better than camera B” means the same as “camera B is worse than camera A”; but the features of these two sentences are completely different. In this paper, we approach comparative opinion classification through prompt learning, taking the advantage of embedded knowledge in pre-trained language model. We design a twin framework with dual prompts, named CORT. This extremely simple model delivers state-of-the-art and robust performance on all benchmark datasets for comparative opinion classification. We believe CORT well serves as a new baseline for comparative opinion classification.

Quotation extraction aims to extract quotations from written text. There are three components in a quotation: source refers to the holder of the quotation, cue is the trigger word(s), and content is the main body. Existing solutions for quotation extraction mainly utilize rule-based approaches and sequence labeling models. While rule-based approaches often lead to low recalls, sequence labeling models cannot well handle quotations with complicated structures. In this paper, we propose the Context and Former-Label Enhanced Net () for quotation extraction. is able to extract complicated quotations with components of variable lengths and complicated structures. On two public datasets (and ) and one proprietary dataset (), we show that our achieves state-of-the-art performance on complicated quotation extraction.

Given an untrimmed video and a text query, natural language video localization (NLVL) is to locate a matching span from the video that semantically corresponds to the query. Existing solutions formulate NLVL either as a ranking task and apply multimodal matching architecture, or as a regression task to directly regress the target video span. In this work, we address NLVL task with a span-based QA approach by treating the input video as text passage. We propose a video span localizing network (VSLNet), on top of the standard span-based QA framework, to address NLVL. The proposed VSLNet tackles the differences between NLVL and span-based QA through a simple and yet effective query-guided highlighting (QGH) strategy. The QGH guides VSLNet to search for matching video span within a highlighted region. Through extensive experiments on three benchmark datasets, we show that the proposed VSLNet outperforms the state-of-the-art methods; and adopting span-based QA framework is a promising direction to solve NLVL.

Cloze-style reading comprehension in Chinese is still limited due to the lack of various corpora. In this paper we propose a large-scale Chinese cloze test dataset ChID, which studies the comprehension of idiom, a unique language phenomenon in Chinese. In this corpus, the idioms in a passage are replaced by blank symbols and the correct answer needs to be chosen from well-designed candidate idioms. We carefully study how the design of candidate idioms and the representation of idioms affect the performance of state-of-the-art models. Results show that the machine accuracy is substantially worse than that of human, indicating a large space for further research.

Biomedical concepts are often mentioned in medical documents under different name variations (synonyms). This mismatch between surface forms is problematic, resulting in difficulties pertaining to learning effective representations. Consequently, this has tremendous implications such as rendering downstream applications inefficacious and/or potentially unreliable. This paper proposes a new framework for learning robust representations of biomedical names and terms. The idea behind our approach is to consider and encode contextual meaning, conceptual meaning, and the similarity between synonyms during the representation learning process. Via extensive experiments, we show that our proposed method outperforms other baselines on a battery of retrieval, similarity and relatedness benchmarks. Moreover, our proposed method is also able to compute meaningful representations for unseen names, resulting in high practical utility in real-world applications.

In multi-document summarization, a set of documents to be summarized is assumed to be on the same topic, known as the underlying topic in this paper. That is, the underlying topic can be collectively represented by all the documents in the set. Meanwhile, different documents may cover various different subtopics and the same subtopic can be across several documents. Inspired by topic model, the underlying topic of a document set can also be viewed as a collection of different subtopics of different importance. In this paper, we propose a summarization model called STDS. The model generates the underlying topic representation from both document view and subtopic view in parallel. The learning objective is to minimize the distance between the representations learned from the two views. The contextual information is encoded through a hierarchical RNN architecture. Sentence salience is estimated in a hierarchical way with subtopic salience and relative sentence salience, by considering the contextual information. Top ranked sentences are then extracted as a summary. Note that the notion of subtopic enables us to bring in additional information (e.g. comments to news articles) that is helpful for document summarization. Experimental results show that the proposed solution outperforms state-of-the-art methods on benchmark datasets.

Extracting time expressions from free text is a fundamental task for many applications. We analyze the time expressions from four datasets and find that only a small group of words are used to express time information, and the words in time expressions demonstrate similar syntactic behaviour. Based on the findings, we propose a type-based approach, named SynTime, to recognize time expressions. Specifically, we define three main syntactic token types, namely time token, modifier, and numeral, to group time-related regular expressions over tokens. On the types we design general heuristic rules to recognize time expressions. In recognition, SynTime first identifies the time tokens from raw text, then searches their surroundings for modifiers and numerals to form time segments, and finally merges the time segments to time expressions. As a light-weight rule-based tagger, SynTime runs in real time, and can be easily expanded by simply adding keywords for the text of different types and of different domains. Experiment on benchmark datasets and tweets data shows that SynTime outperforms state-of-the-art methods.