COLING 2024 论文列表

Language identification is a crucial component in the automated production of language resources, particularly in multilingual and big data contexts. However, commonly used language identifiers struggle to differentiate between similar or closely-related languages. This paper introduces FastSpell, a language identifier that combines fastText (a pre-trained language identifier tool) and Hunspell (a spell checker) with the aim of having a refined second-opinion before deciding which language should be assigned to a text. We provide a description of the FastSpell algorithm along with an explanation on how to use and configure it. To that end, we motivate the need of such a tool and present a benchmark including some popular language identifiers evaluated during the development of FastSpell. We show how FastSpell is useful not only to improve identification of similar languages, but also to identify new ones ignored by other tools.

Large language models (LLMs) have achieved great success in a variety of natural language understanding tasks. However, domain discrepancies between the downstream task and the pre-training corpora may have hurdled LLMs to excel further in the vertical applications. Contrary to prior computational-heavy methods, we propose a lightweight solution to further bridge the gap in applying LLMs to diverse downstream tasks — a Fast Adaptation method for LLMs via Prompted Data, in short FAvPD. Notably, with FAvPD, we establish an additional adaptive tuning procedure, wherein we integrate downstream text corpora, gold labels as well as external knowledge sources and then envelop them into a form of highly controllable prompt. As a simple, easy-to-use, and versatile solution, FAvPD lies in the intersection of regimes like knowledge-augmented LLMs, fine-tuning, and adaptation techniques. With extensive experiments, we prove that FAvPD excels in both performance efficacy and training efficiency over related prior works. FAvPD is publicly available at https://github.com/Hyatio/FAvPD.

End-to-end (E2E) Spoken Language Understanding (SLU) systems infer structured information directly from the speech signal using a single model. Due to the success of virtual assistants and the increasing demand for speech interfaces, these architectures are being actively researched for their potential to improve system performance by exploiting acoustic information and avoiding the cascading errors of traditional architectures. However, these systems require large amounts of specific, well-labelled speech data for training, which is expensive to obtain even in English, where the number of public audio datasets for SLU is limited. In this paper, we release the FalAI dataset, the largest public SLU dataset in terms of hours (250 hours), recordings (260,000) and participants (over 10,000), which is also the first SLU dataset in Galician and the first to be obtained in a low-resource scenario. Furthermore, we present new measures of complexity for the text corpora, the strategies followed for the design, collection and validation of the dataset, and we define splits for noisy audio, hesitant audio and audio where the sentence has changed but the structured information is preserved. These novel splits provide a unique resource for testing SLU systems in challenging, real-world scenarios.

Multi-domain aspect-based sentiment analysis (ABSA) seeks to capture fine-grained sentiment across diverse domains. While existing research narrowly focuses on single-domain applications constrained by methodological limitations and data scarcity, the reality is that sentiment naturally traverses multiple domains. Although large language models (LLMs) offer a promising solution for ABSA, it is difficult to integrate effectively with established techniques, including graph-based models and linguistics, because modifying their internal architecture is not easy. To alleviate this problem, we propose a novel framework, Feature-aware In-context Learning for Multi-domain ABSA (FaiMA). The core insight of FaiMA is to utilize in-context learning (ICL) as a feature-aware mechanism that facilitates adaptive learning in multi-domain ABSA tasks. Specifically, we employ a multi-head graph attention network as a text encoder optimized by heuristic rules for linguistic, domain, and sentiment features. Through contrastive learning, we optimize sentence representations by focusing on these diverse features. Additionally, we construct an efficient indexing mechanism, allowing FaiMA to stably retrieve highly relevant examples across multiple dimensions for any given input. To evaluate the efficacy of FaiMA, we build the first multi-domain ABSA benchmark dataset. Extensive experimental results demonstrate that FaiMA achieves significant performance improvements in multiple domains compared to baselines, increasing F1 by 2.07% on average. Source code and data sets are available at https://github.com/SupritYoung/FaiMA.

Previous Sign Language Translation (SLT) methods achieve superior performance by relying on gloss annotations. However, labeling high-quality glosses is a labor-intensive task, which limits the further development of SLT. Although some approaches work towards gloss-free SLT through jointly training the visual encoder and translation network, these efforts still suffer from poor performance and inefficient use of the powerful Large Language Model (LLM). Most seriously, we find that directly introducing LLM into SLT will lead to insufficient learning of visual representations as LLM dominates the learning curve. To address these problems, we propose Factorized Learning assisted with Large Language Model (FLa-LLM) for gloss-free SLT. Concretely, we factorize the training process into two stages. In the visual initialing stage, we employ a lightweight translation model after the visual encoder to pre-train the visual encoder. In the LLM fine-tuning stage, we freeze the acquired knowledge in the visual encoder and integrate it with a pre-trained LLM to inspire the LLM’s translation potential. This factorized training strategy proves to be highly effective as evidenced by significant improvements achieved across three SLT datasets which are all conducted under the gloss-free setting.

In the face of the rapidly growing spread of false and misleading information in the real world, manual evidence-based fact-checking efforts become increasingly challenging and time-consuming. In order to tackle this issue, we propose FaGANet, an automated and accurate fact-checking model that leverages the power of sentence-level attention and graph attention network to enhance performance. This model adeptly integrates encoder-only models with graph attention network, effectively fusing claims and evidence information for accurate identification of even well-disguised data. Experiment results showcase the significant improvement in accuracy achieved by our FaGANet model, as well as its state-of-the-art performance in the evidence-based fact-checking task. We release our code and data in https://github.com/WeiyaoLuo/FaGANet.

This study introduces a novel multimodal corpus for expressive task-based spoken language and dialogue, focused on language use under frustration and surprise, elicited from three tasks motivated by prior research and collected in an IRB-approved experiment. The resource is unique both because these are understudied affect states for emotion modeling in language, and also because it provides both individual and dyadic multimodally grounded language. The study includes a detailed analysis of annotations and performance results for multimodal emotion inference in language use.

FReND is a freely available corpus of French language in which negations are hand-annotated. Negations are annotated by their cues and scopes. Comprising 590K tokens and over 8.9K negations, it is the largest dataset available for French. A variety of types of textual genres are covered: literature, blog posts, Wikipedia articles, political debates, clinical reports and newspaper articles. As the understanding of negation is not yet mastered by current state of the art AI-models, FReND is not only a valuable resource for linguistic research into negation, but also as training data for AI tasks such as negation detection.

Natural language processing (NLP) applications such as named entity recognition (NER) for low-resource corpora do not benefit from recent advances in the development of large language models (LLMs) where there is still a need for larger annotated datasets. This research article introduces a methodology for generating translated versions of annotated datasets through crosslingual annotation projection and is freely available on GitHub (link: https://github.com/JamilProg/crosslingual_bert_annotation_projection). Leveraging a language agnostic BERT-based approach, it is an efficient solution to increase low-resource corpora with few human efforts and by only using already available open data resources. Quantitative and qualitative evaluations are often lacking when it comes to evaluating the quality and effectiveness of semi-automatic data generation strategies. The evaluation of our crosslingual annotation projection approach showed both effectiveness and high accuracy in the resulting dataset. As a practical application of this methodology, we present the creation of French Annotated Resource with Semantic Information for Medical Entities Detection (FRASIMED), an annotated corpus comprising 2’051 synthetic clinical cases in French. The corpus is now available for researchers and practitioners to develop and refine French natural language processing (NLP) applications in the clinical field (https://zenodo.org/record/8355629), making it the largest open annotated corpus with linked medical concepts in French.

Quotation extraction is a widely useful task both from a sociological and from a Natural Language Processing perspective. However, very little data is available to study this task in languages other than English. In this paper, we present FRACAS, a manually annotated corpus of 1,676 newswire texts in French for quotation extraction and source attribution. We first describe the composition of our corpus and the choices that were made in selecting the data. We then detail the annotation guidelines, the annotation process and give relevant statistics about our corpus. We give results for the inter-annotator agreement, which is substantially high for such a difficult linguistic phenomenon. We use this new resource to test the ability of a neural state-of-the-art relation extraction system to extract quotes and their source and we compare this model to the latest available system for quotation extraction for the French language, which is rule-based. Experiments using our dataset on the state-of-the-art system show very promising results considering the difficulty of the task at hand.

It is known from large-scale crowd experimentation that some people are innately better at analysing complex situations and making justified predictions – the so-called ‘superforecasters’. Surprisingly, however, there has to date been no work exploring the role played by the reasoning in those justifications. Bag-of-words analyses might tell us something, but the real value lies in understanding what features of reasoning and argumentation lead to better forecasts – both in providing an objective measure for argument quality, and even more importantly, in providing guidance on how to improve forecasting performance. The work presented here covers the creation of a unique dataset of such prediction rationales, the structure of which naturally lends itself to partially automated annotation which in turn is used as the basis for subsequent manual enhancement that provides a uniquely fine-grained and close characterisation of the structure of argumentation, with potential impact on forecasting domains from intelligence analysis to investment decision-making.

Large language models have amply proven their great capabilities, both in downstream tasks and real-life settings. However, low- and mid-resource languages do not have access to the necessary means to train such models from scratch, and often have to rely on multilingual models despite being underrepresented in the training data. For the particular case of the Catalan language, we prove that continued pre-training with vocabulary adaptation is a better alternative to take the most out of already pre-trained models, even if these have not seen any Catalan data during their pre-training phase. We curate a 26B tokens corpus and use it to further pre-train BLOOM, giving rise to the FLOR models. We perform an extensive evaluation to assess the effectiveness of our method, obtaining consistent gains across Catalan and Spanish tasks. The models, training data, and evaluation framework are made freely available under permissive licenses.

In this paper, we introduce the Fongbe to French Speech Translation Corpus (FFSTC). This corpus encompasses approximately 31 hours of collected Fongbe language content, featuring both French transcriptions and corresponding Fongbe voice recordings. FFSTC represents a comprehensive dataset compiled through various collection methods and the efforts of dedicated individuals. Furthermore, we conduct baseline experiments using Fairseq’s transformer_s and conformer models to evaluate data quality and validity. Our results indicate a score BLEU of 8.96 for the transformer_s model and 8.14 for the conformer model, establishing a baseline for the FFSTC corpus.

Document-level Relation Extraction (DocRE) aims to identify relation labels between entities within a single document. It requires handling several sentences and reasoning over them. State-of-the-art DocRE methods use a graph structure to connect entities across the document to capture dependency syntax information. However, this is insufficient to fully exploit the rich syntax information in the document. In this work, we propose to fuse constituency and dependency syntax into DocRE. It uses constituency syntax to aggregate the whole sentence information and select the instructive sentences for the pairs of targets. It exploits dependency syntax in a graph structure with constituency syntax enhancement and chooses the path between entity pairs based on the dependency graph. The experimental results on datasets from various domains demonstrate the effectiveness of the proposed method.

LeiLanD (Leiden Language Data) is a searchable catalogue initiated by the Leiden University Centre for Linguistics (LUCL) with the support of CLARIAH. The catalogue contains metadata about language datasets collected at LUCL and other institutes of Leiden University. This paper describes a project to FAIRify the datasets increasing their findability and accessibility through a standardised metadata format CMDI so as to obtain a rich metadata description for all resources and to make them findable through CLARIN’s Virtual Language Observatory. The paper describes the creation of the catalogue and the steps that led from unstructured metadata to CMDI standards. This FAIRifi- cation of LeiLanD has enhanced the findability and accessibility of incredibly diverse collection of language datasets.

Eye movement features are considered to be direct signals reflecting human attention distribution with a low cost to obtain, inspiring researchers to augment language models with eye-tracking (ET) data. In this study, we select first fixation duration (FFD) and total reading time (TRT) as the cognitive signals to guide Transformer attention in question-answering (QA) tasks. We design three different ET attention masks based on the two features, either collected from human reading events or generated by a gaze-predicting model. We augment BERT and ALBERT models with attention masks structured based on the ET data. We find that augmenting a model with ET data carries linguistic features complementing the information captured by the model. It improves the models’ performance but compromises the stability. Different Transformer models benefit from different types of ET attention masks, while ALBERT performs better than BERT. Moreover, ET data collected from real-life reading events has better model augmenting ability than the model-predicted data.

Social determinants of health (SDoH) play a critical role in shaping health outcomes, particularly in pediatric populations where interventions can have long-term implications. SDoH are frequently studied in the Electronic Health Record (EHR), which provides a rich repository for diverse patient data. In this work, we present a novel annotated corpus, the Pediatric Social History Annotation Corpus (PedSHAC), and evaluate the automatic extraction of detailed SDoH representations using fine-tuned and in-context learning methods with Large Language Models (LLMs). PedSHAC comprises annotated social history sections from 1,260 clinical notes obtained from pediatric patients within the University of Washington (UW) hospital system. Employing an event-based annotation scheme, PedSHAC captures ten distinct health determinants to encompass living and economic stability, prior trauma, education access, substance use history, and mental health with an overall annotator agreement of 81.9 F1. Our proposed fine-tuning LLM-based extractors achieve high performance at 78.4 F1 for event arguments. In-context learning approaches with GPT-4 demonstrate promise for reliable SDoH extraction with limited annotated examples, with extraction performance at 82.3 F1 for event triggers.

Event Extraction (EE) is widely used in the Chinese financial field to provide valuable structured information. However, there are two key challenges for Chinese financial EE in application scenarios. First, events need to be extracted from raw texts, which sets it apart from previous works like the Automatic Content Extraction (ACE) EE task, where EE is treated as a classification problem given the entity spans. Second, recognizing financial entities can be laborious, as they may involve multiple elements. In this paper, we introduce CFTE, a novel task for Chinese Financial Text-to-Event extraction, which directly extracts financial events from raw texts. We further present FINEED, a Chinese FINancial Event Extraction Dataset, and an efficient MAtrix-ChunKing method called MACK, designed for the extraction of financial events from raw texts. Specifically, FINEED is manually annotated with rich linguistic features. We propose a novel two-dimensional annotation method for FINEED, which can visualize the interactions among text components. Our MACK method is fault-tolerant by preserving the tag frequency distribution when identifying financial entities. We conduct extensive experiments and the results verify the effectiveness of our MACK method.

Automatic Speech Recognition (ASR) technology is fundamental in transcribing spoken language into text, with considerable applications in the clinical realm, including streamlining medical transcription and integrating with Electronic Health Record (EHR) systems. Nevertheless, challenges persist, especially when transcriptions contain noise, leading to significant drops in performance when Natural Language Processing (NLP) models are applied. Named Entity Recognition (NER), an essential clinical task, is particularly affected by such noise, often termed the ASR-NLP gap. Prior works have primarily studied ASR’s efficiency in clean recordings, leaving a research gap concerning the performance in noisy environments. This paper introduces a novel dataset, BioASR-NER, designed to bridge the ASR-NLP gap in the biomedical domain, focusing on extracting adverse drug reactions and mentions of entities from the Brief Test of Adult Cognition by Telephone (BTACT) exam. Our dataset offers a comprehensive collection of almost 2,000 clean and noisy recordings. In addressing the noise challenge, we present an innovative transcript-cleaning method using GPT-4, investigating both zero-shot and few-shot methodologies. Our study further delves into an error analysis, shedding light on the types of errors in transcription software, corrections by GPT-4, and the challenges GPT-4 faces. This paper aims to foster improved understanding and potential solutions for the ASR-NLP gap, ultimately supporting enhanced healthcare documentation practices.