Proceedings of the conference. Association for Computational Linguistics. Meeting最新文献

IRIS: Interpretable Retrieval-Augmented Classification for Long Interspersed Document Sequences. IRIS：长分散文档序列的可解释检索增强分类。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2025-07-01

Fengnan Li, Elliot D Hill, Shu Jiang, Jiaxin Gao, Matthew M Engelhard

{"title":"IRIS: Interpretable Retrieval-Augmented Classification for Long Interspersed Document Sequences.","authors":"Fengnan Li, Elliot D Hill, Shu Jiang, Jiaxin Gao, Matthew M Engelhard","doi":"","DOIUrl":"","url":null,"abstract":"Transformer-based models have achieved state-of-the-art performance in document classification but struggle with long-text processing due to the quadratic computational complexity in the self-attention module. Existing solutions, such as sparse attention, hierarchical models, and key sentence extraction, partially address the issue but still fall short when the input sequence is exceptionally lengthy. To address this challenge, we propose IRIS (Interpretable Retrieval-Augmented Classification for long Interspersed Document Sequences), a novel, lightweight framework that utilizes retrieval to efficiently classify long documents while enhancing interpretability. IRIS segments documents into chunks, stores their embeddings in a vector database, and retrieves those most relevant to a given task using learnable query vectors. A linear attention mechanism then aggregates the retrieved embeddings for classification, allowing the model to process arbitrarily long documents without increasing computational cost and remaining trainable on a single GPU. Our experiments across six datasets show that IRIS achieves comparable performance to baseline models on standard benchmarks, and excels in three clinical note disease risk prediction tasks where documents are extremely long and key information is sparse. Furthermore, IRIS provides global interpretability by revealing a clear summary of key risk factors identified by the model. These findings highlight the potential of IRIS as an efficient and interpretable solution for long-document classification, particularly in healthcare applications where both performance and explainability are crucial.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2025 ","pages":"30263-30283"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12357761/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144877195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

GraphCheck: Breaking Long-Term Text Barriers with Extracted Knowledge Graph-Powered Fact-Checking. GraphCheck：通过提取知识图驱动的事实检查打破长期文本障碍。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2025-07-01 DOI: 10.18653/v1/2025.acl-long.729

Yingjian Chen, Haoran Liu, Yinhong Liu, Jinxiang Xie, Rui Yang, Han Yuan, Yanran Fu, Peng Yuan Zhou, Qingyu Chen, James Caverlee, Irene Li

{"title":"GraphCheck: Breaking Long-Term Text Barriers with Extracted Knowledge Graph-Powered Fact-Checking.","authors":"Yingjian Chen, Haoran Liu, Yinhong Liu, Jinxiang Xie, Rui Yang, Han Yuan, Yanran Fu, Peng Yuan Zhou, Qingyu Chen, James Caverlee, Irene Li","doi":"10.18653/v1/2025.acl-long.729","DOIUrl":"10.18653/v1/2025.acl-long.729","url":null,"abstract":"Large language models (LLMs) are widely used, but they often generate subtle factual errors, especially in long-form text. These errors are fatal in some specialized domains such as medicine. Existing fact-checking with grounding documents methods face two main challenges: (1) they struggle to understand complex multihop relations in long documents, often overlooking subtle factual errors; (2) most specialized methods rely on pairwise comparisons, requiring multiple model calls, leading to high resource and computational costs. To address these challenges, we propose GraphCheck , a fact-checking framework that uses extracted knowledge graphs to enhance text representation. Graph Neural Networks further process these graphs as a soft prompt, enabling LLMs to incorporate structured knowledge more effectively. Enhanced with graph-based reasoning, GraphCheck captures multihop reasoning chains that are often overlooked by existing methods, enabling precise and efficient fact-checking in a single inference call. Experimental results on seven benchmarks spanning both general and medical domains demonstrate up to a 7.1% overall improvement over baseline models. Notably, GraphCheck outperforms existing specialized fact-checkers and achieves comparable performance with state-of-the-art LLMs, such as DeepSeek-V3 and OpenAI-o1, with significantly fewer parameters.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2025 ","pages":"14976-14995"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12360635/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144884476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

OLIVE: Object Level In-Context Visual Embeddings. OLIVE：对象级上下文视觉嵌入。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2024-08-01 DOI: 10.18653/v1/2024.acl-long.282

Timothy Ossowski, Junjie Hu

{"title":"OLIVE: Object Level In-Context Visual Embeddings.","authors":"Timothy Ossowski, Junjie Hu","doi":"10.18653/v1/2024.acl-long.282","DOIUrl":"10.18653/v1/2024.acl-long.282","url":null,"abstract":"Recent generalist vision-language models (VLMs) have demonstrated impressive reasoning capabilities across diverse multimodal tasks. However, these models still struggle with fine-grained object level understanding and grounding. In terms of modeling, existing VLMs implicitly align text tokens with image patch tokens, which is ineffective for embedding alignment at the same granularity and inevitably introduces noisy spurious background features. Additionally, these models struggle when generalizing to unseen visual concepts and may not be reliable for domain-specific tasks without further fine-tuning. To address these limitations, we propose a novel method to prompt large language models with in-context visual object vectors, thereby enabling controllable object level reasoning. This eliminates the necessity of fusing a lengthy array of image patch features and significantly speeds up training. Furthermore, we propose region-level retrieval using our object representations, facilitating rapid adaptation to new objects without additional training. Our experiments reveal that our method achieves competitive referring object classification and captioning performance, while also offering zero-shot generalization and robustness to visually challenging contexts.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2024 ","pages":"5170-5185"},"PeriodicalIF":0.0,"publicationDate":"2024-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11931571/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143702203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unity in Diversity: Collaborative Pre-training Across Multimodal Medical Sources. 多样性中的统一：跨多模式医学资源的协同预训练。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2024-08-01 DOI: 10.18653/v1/2024.acl-long.199

Xiaochen Wang, Junyu Luo, Jiaqi Wang, Yuan Zhong, Xiaokun Zhang, Yaqing Wang, Parminder Bhatia, Cao Xiao, Fenglong Ma

{"title":"Unity in Diversity: Collaborative Pre-training Across Multimodal Medical Sources.","authors":"Xiaochen Wang, Junyu Luo, Jiaqi Wang, Yuan Zhong, Xiaokun Zhang, Yaqing Wang, Parminder Bhatia, Cao Xiao, Fenglong Ma","doi":"10.18653/v1/2024.acl-long.199","DOIUrl":"https://doi.org/10.18653/v1/2024.acl-long.199","url":null,"abstract":"Although pre-training has become a prevalent approach for addressing various biomedical tasks, the current efficacy of pre-trained models is hindered by their reliance on a limited scope of medical sources. This limitation results in data scarcity during pre-training and restricts the range of applicable downstream tasks. In response to these challenges, we develop Medical Cross-Source Pre-training (MEDCSP), a new pre-training strategy designed to bridge the gap between multimodal medical sources. MEDCSP employs modality-level aggregation to unify patient data within individual sources. Additionally, leveraging temporal information and diagnosis history, MEDCSP effectively captures explicit and implicit correlations between patients across different sources. To evaluate the proposed strategy, we conduct comprehensive experiments, where the experiments are based on 6 modalities from 2 real-world medical data sources, and MEDCSP is evaluated on 4 tasks against 19 baselines, marking an initial yet essential step towards cross-source modeling in the medical domain.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2024 Volume 1 Long Papers","pages":"3644-3656"},"PeriodicalIF":0.0,"publicationDate":"2024-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12007664/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144054485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Medical Vision-Language Pre-Training for Brain Abnormalities. 针对大脑异常的医学视觉语言预培训。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2024-05-01

Masoud Monajatipoor, Zi-Yi Dou, Aichi Chien, Nanyun Peng, Kai-Wei Chang

{"title":"Medical Vision-Language Pre-Training for Brain Abnormalities.","authors":"Masoud Monajatipoor, Zi-Yi Dou, Aichi Chien, Nanyun Peng, Kai-Wei Chang","doi":"","DOIUrl":"","url":null,"abstract":"Vision-language models have become increasingly powerful for tasks that require an understanding of both visual and linguistic elements, bridging the gap between these modalities. In the context of multimodal clinical AI, there is a growing need for models that possess domain-specific knowledge, as existing models often lack the expertise required for medical applications. In this paper, we take brain abnormalities as an example to demonstrate how to automatically collect medical image-text aligned data for pretraining from public resources such as PubMed. In particular, we present a pipeline that streamlines the pre-training process by initially collecting a large brain image-text dataset from case reports and published journals and subsequently constructing a high-performance vision-language model tailored to specific medical tasks. We also investigate the unique challenge of mapping subfigures to subcaptions in the medical domain. We evaluated the resulting model with quantitative and qualitative intrinsic evaluations. The resulting dataset and our code can be found here https://github.com/masoud-monajati/MedVL_pretraining_pipeline.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2024 LREC/COLING","pages":"11159-11164"},"PeriodicalIF":0.0,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11238846/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141617775","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

HiGen: Hierarchy-Aware Sequence Generation for Hierarchical Text Classification. 层次文本分类的层次感知序列生成。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2024-03-01

Vidit Jain, Mukund Rungta, Yuchen Zhuang, Yue Yu, Zeyu Wang, Mu Gao, Jeffrey Skolnick, Chao Zhang

{"title":"HiGen: Hierarchy-Aware Sequence Generation for Hierarchical Text Classification.","authors":"Vidit Jain, Mukund Rungta, Yuchen Zhuang, Yue Yu, Zeyu Wang, Mu Gao, Jeffrey Skolnick, Chao Zhang","doi":"","DOIUrl":"","url":null,"abstract":"Hierarchical text classification (HTC) is a complex subtask under multi-label text classification, characterized by a hierarchical label taxonomy and data imbalance. The best-performing models aim to learn a static representation by combining document and hierarchical label information. However, the relevance of document sections can vary based on the hierarchy level, necessitating a dynamic document representation. To address this, we propose HiGen, a text-generation-based framework utilizing language models to encode dynamic text representations. We introduce a level-guided loss function to capture the relationship between text and label name semantics. Our approach incorporates a task-specific pretraining strategy, adapting the language model to in-domain knowledge and significantly enhancing performance for classes with limited examples. Furthermore, we present a new and valuable dataset called ENZYME, designed for HTC, which comprises articles from PubMed with the goal of predicting Enzyme Commission (EC) numbers. Through extensive experiments on the ENZYME dataset and the widely recognized WOS and NYT datasets, our methodology demonstrates superior performance, surpassing existing approaches while efficiently handling data and mitigating class imbalance. We release our code and dataset here: https://github.com/viditjain99/HiGen.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2024 EACL","pages":"1354-1368"},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11781299/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143070252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improving the Transferability of Clinical Note Section Classification Models with BERT and Large Language Model Ensembles. 利用BERT和大型语言模型集成提高临床笔记部分分类模型的可移植性。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2023-07-01

Weipeng Zhou, Dmitriy Dligach, Majid Afshar, Yanjun Gao, Timothy A Miller

引用次数: 0

Improving the Transferability of Clinical Note Section Classification Models with BERT and Large Language Model Ensembles 利用BERT和大型语言模型集成提高临床笔记部分分类模型的可移植性

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2023-07-01 DOI: 10.18653/v1/2023.clinicalnlp-1.16

Weipeng Zhou, M. Afshar, Dmitriy Dligach, Yanjun Gao, Timothy Miller

引用次数: 0

Less Likely Brainstorming: Using Language Models to Generate Alternative Hypotheses. 不太可能的头脑风暴:使用语言模型生成替代假设。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2023-07-01 DOI: 10.18653/v1/2023.findings-acl.794

Liyan Tang, Yifan Peng, Yanshan Wang, Ying Ding, Greg Durrett, Justin F Rousseau

{"title":"Less Likely Brainstorming: Using Language Models to Generate Alternative Hypotheses.","authors":"Liyan Tang, Yifan Peng, Yanshan Wang, Ying Ding, Greg Durrett, Justin F Rousseau","doi":"10.18653/v1/2023.findings-acl.794","DOIUrl":"https://doi.org/10.18653/v1/2023.findings-acl.794","url":null,"abstract":"A human decision-maker benefits the most from an AI assistant that corrects for their biases. For problems such as generating interpretation of a radiology report given findings, a system predicting only highly likely outcomes may be less useful, where such outcomes are already obvious to the user. To alleviate biases in human decision-making, it is worth considering a broad differential diagnosis, going beyond the most likely options. We introduce a new task, \"less likely brainstorming,\" that asks a model to generate outputs that humans think are relevant but less likely to happen. We explore the task in two settings: a brain MRI interpretation generation setting and an everyday commonsense reasoning setting. We found that a baseline approach of training with less likely hypotheses as targets generates outputs that humans evaluate as either likely or irrelevant nearly half of the time; standard MLE training is not effective. To tackle this problem, we propose a controlled text generation method that uses a novel contrastive learning strategy to encourage models to differentiate between generating likely and less likely outputs according to humans. We compare our method with several state-of-the-art controlled text generation models via automatic and human evaluations and show that our models' capability of generating less likely outputs is improved.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2023 ","pages":"12532-12555"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10494958/pdf/nihms-1923571.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10263511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Revisiting Relation Extraction in the era of Large Language Models. 再论大语言模型时代的关系抽取。

Proceedings of the conference. Association for Computational Linguistics. Meeting Pub Date : 2023-07-01 DOI: 10.18653/v1/2023.acl-long.868

Somin Wadhwa, Silvio Amir, Byron C Wallace

{"title":"Revisiting Relation Extraction in the era of Large Language Models.","authors":"Somin Wadhwa, Silvio Amir, Byron C Wallace","doi":"10.18653/v1/2023.acl-long.868","DOIUrl":"https://doi.org/10.18653/v1/2023.acl-long.868","url":null,"abstract":"Relation extraction (RE) is the core NLP task of inferring semantic relationships between entities from text. Standard supervised RE techniques entail training modules to tag tokens comprising entity spans and then predict the relationship between them. Recent work has instead treated the problem as a sequence-to-sequence task, linearizing relations between entities as target strings to be generated conditioned on the input. Here we push the limits of this approach, using larger language models (GPT-3 and Flan-T5 large) than considered in prior work and evaluating their performance on standard RE tasks under varying levels of supervision. We address issues inherent to evaluating generative approaches to RE by doing human evaluations, in lieu of relying on exact matching. Under this refined evaluation, we find that: (1) Few-shot prompting with GPT-3 achieves near SOTA performance, i.e., roughly equivalent to existing fully supervised models; (2) Flan-T5 is not as capable in the few-shot setting, but supervising and fine-tuning it with Chain-of-Thought (CoT) style explanations (generated via GPT-3) yields SOTA results. We release this model as a new baseline for RE tasks.","PeriodicalId":74541,"journal":{"name":"Proceedings of the conference. Association for Computational Linguistics. Meeting","volume":"2023 ","pages":"15566-15589"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10482322/pdf/nihms-1912166.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10181357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0