Austria

We describe our system for the ArchEHR-QA Shared Task on answering clinical questions using electronic health records (EHRs). Our approach uses large language models in two steps: first, to find sentences in the EHR relevant to a clinician&#39;s question, and second, to generate a short, citation-supported response based on those sentences. We use few-shot prompting, self-consistency, and thresholding to improve the sentence classification step to decide which sentences are essential. We compare several models and find that a smaller 8B model performs better than a larger 70B model for identifying relevant information. Our results show that accurate sentence selection is critical for generating high-quality responses and that self-consistency with thresholding helps make these decisions more reliable.

ACL 2025

UTSA-NLP at ArchEHR-QA 2025: Improving EHR Question Answering via Self-Consistency Prompting

We describe our system for the ArchEHR-QA Shared Task on answering clinical questions using electronic health records (EHRs). Our approach uses large language models in two steps: first, to find sentences in the EHR relevant to a clinician's question, and second, to generate a short, citation-supported response based on those sentences. We use few-shot prompting, self-consistency, and thresholding to improve the sentence classification step to decide which sentences are essential. We compare several models and find that a smaller 8B model performs better than a larger 70B model for identifying relevant information. Our results show that accurate sentence selection is critical for generating high-quality responses and that self-consistency with thresholding helps make these decisions more reliable.

workshop paper

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Biomedical texts, such as research articles and clinical reports, are often written in highly technical language, making them difficult for patients and the general public to understand. The BioLaySumm 2025 Shared Task addresses this challenge by promoting the development of models that generate lay summarisation of biomedical content. This paper focuses on Subtask 2.1: Radiology Report Generation with Layman's Terms. In this work, we evaluate two large language model (LLM) architectures, T5-large (700M parameter encoder–decoder model) and LLaMA-3.2-3B (3B parameter decoder-only model). Both models are trained under fully-supervised conditions using the task's multi-source dataset. Our results show that T5-large consistently outperforms LLaMA-3.2-3B across nine out of ten metrics, including relevance, readability, and clinical accuracy, despite having only a quarter of the parameters. Our T5-based model achieved the top rank in both the open-source and close-source tracks of the subtask 2.1.

AEHRC at BioLaySumm 2025: Leveraging T5 for Lay Summarisation of Radiology Reports

In this paper, we present our approach to the BioLaySumm 2025 Shared Task on lay summarization of biomedical research articles, which was conducted as part of the BioNLP Workshop 2025. The aim of the task is to create lay summaries from scientific articles to improve accessibility for a non-expert audience. To this end, we applied preprocessing techniques to clean and standardize the input texts, and fine-tuned Qwen2.5 and Qwen3-based language models for the summarization task. For abstract-based fine-tuning, we investigated whether we can insert salient sentences from the main article into the summary to enrich the input. We also curated a dataset of child-friendly articles with corresponding gold-standard summaries and used large language models to rewrite them into more complex scientific variants to augment our training data with more examples.

MetninOzU at BioLaySumm2025: Text Summarization with Reverse Data Augmentation and Injecting Salient Sentences

Biomedical articles are often inaccessible to non-experts due to their technical complexity. To improve readability and factuality of lay summaries, we built on an extract-then-summarize framework by experimenting with novel extractive summarization strategies and employing Low Rank Adaptation (LoRA) fine-tuning of Meta-Llama-3-8B-Instruct on data selected by these strategies. We also explored counterfactual data augmentation and post-processing definition insertion to further enhance factual grounding and accessibility. Our best performing system treats the article's title and keywords (i.e. search terms) as a single semantic centroid and ranks sentences by their semantic similarity to this centroid. This constrained selection of data serves as input for fine-tuning, achieving marked improvements in readability and factuality of downstream abstractive summaries while maintaining relevance. Our approach highlights the importance of quality data curation for biomedical lay summarization, resulting in 4th best overall performance and 2nd best Readability performance for the BioLaySumm 2025 Shared Task at BioNLP 2025.

MIRAGES at BioLaySumm2025: The Impact of Search Terms and Data Curation for Biomedical Lay Summarization

We revisit the BARTABSA framework for aspect-based sentiment analysis with modern decoder LLMs to assess the importance of explicit structure modeling today. Our updated implementation - BARTABSA++ - features architectural enhancements that boost performance and training stability.
Systematic testing with various encoder-decoder architectures shows that BARTABSA++ with BART-Large achieves state-of-the-art results, even surpassing a finetuned GPT-4o model.
Our analysis indicates the encoder's representational quality is vital, while the decoder's role is minimal, explaining the limited benefits of scaling decoder-only LLMs for this task. These findings underscore the complementary roles of explicit structured modeling and large language models, indicating structured approaches remain competitive for tasks requiring precise relational information extraction.

BARTABSA++: Revisiting BARTABSA with Decoder LLMs

In this paper, we explore the use of a text-only, autoregressive language modeling approach for the extraction of referring expressions from visually grounded dialogue. More specifically, the aim is to investigate the extent to which the linguistic context alone can inform the detection of mentions that have a (visually perceivable) referent in the visual context of the conversation. To this end, we adapt a pretrained large language model (LLM) to perform a relatively course-grained annotation of mention spans in unfolding conversations by demarcating mention span boundaries in text via next-token prediction. Our findings indicate that even when using a moderately sized LLM, relatively small datasets, and parameter-efficient fine-tuning, a text-only approach can be effective, highlighting the relative importance of the linguistic context for this task. Nevertheless, we argue that the task represents an inherently multimodal problem and discuss limitations fundamental to unimodal approaches.

Detecting Referring Expressions in Visually Grounded Dialogue with Autoregressive Language Models

This paper evaluates the ability of Large Language Models (LLMs) to leverage contextual information in the form of structured linguistic representations. Specifically, we examine the impact of encoding both short and long contexts using Abstract Meaning Representation (AMR) structures across a diverse set of language tasks. We perform our analysis using 8-bit quantized and instruction-tuned versions of Llama 3.1 (8B), Phi-3, and Mistral 7B. Our results indicate that, for tasks involving short contexts, augmenting the prompt with the AMR of the original language context often degrades the performance of the underlying LLM. However, for tasks that involve long contexts, such as dialogue summarization in the SAMSum dataset, this enhancement improves LLM performance, for example, by increasing the zero-shot cosine similarity score of Llama 3.1 from 66% to 76%. This improvement is more evident in the newer and larger LLMs, but does not extend to the older or smaller ones. In addition, we observe that LLMs can effectively reconstruct the original text from a linearized AMR, achieving a cosine similarity of 81% in the best-case scenario.

Can LLMs Interpret and Leverage Structured Linguistic Representations? A Case Study with AMRs

Large Language Models (LLMs) have brought significant breakthroughs across all areas of Natural Language Processing (NLP), including Information Extraction (IE). However, knowledge gaps remain regarding their effectiveness in extracting entity-relation triplets, i.e. Joint Relation Extraction (JRE). JRE has been a key operation in creating knowledge bases that can be used to enhance Retrieval Augmented Generation (RAG) systems. Prior work highlights low-quality triplets generated by LLMs. Thus, this work investigates the impact of incorporating linguistic structures, such as constituency and dependency trees and semantic role labeling, to enhance the quality of the extracted triplets. The findings suggest that incorporating specific structural information enhances the uniqueness and topical relevance of the triplets, particularly in scenarios where multiple relationships are present.

From Syntax to Semantics: Evaluating the Impact of Linguistic Structures on LLM-Based Information Extraction

Event-keyed summarization (EKS) requires summarizing a specific event described in a document given the document text and an event representation extracted from it. In this work, we extend EKS to the cross-document setting (CDEKS), in which summaries must synthesize information from accounts of the same event as given by multiple sources. We introduce **SEAMuS** (**S**ummaries of **E**vents **A**cross **Mu**ltiple **S**ources), a high-quality dataset for CDEKS based on an expert reannotation of the FAMuS dataset for cross-document argument extraction. We present a suite of baselines on SEAMuS—covering both smaller, fine-tuned models, as well as zero- and few-shot prompted LLMs—along with detailed ablations and a human evaluation study, showing SEAMuS to be a valuable benchmark for this new task.

Cross-Document Event-Keyed Summarization

Information Extraction (IE) plays a pivotal role in transforming unstructured data into structured formats, such as Knowledge Graphs. One of the main tasks within IE is Relation Extraction (RE), which identifies relations between entities in text data. This process enriches the semantic understanding of documents, enabling more precise information retrieval and query answering. Recent works leveraging pre-trained language models have demonstrated significant performance improvements in RE. In the current era of Large Language Models (LLMs), fine-tuning these LLMs can mitigate the limitations of zero-shot RE methods, particularly in overcoming the domain adaptation challenges inherent in RE. This work explores not only the effectiveness of fine-tuned LLMs but also their integration into a Retrieval-Augmented Generation (RAG)-based RE approach to address domain adaptation challenges when general-purpose LLMs serve as generators within the RAG framework. Empirical evaluations on the TACRED, TACRED-Revisited (TACREV), and Re-TACRED datasets reveal substantial performance improvements with fine-tuned LLMs, such as Llama2-7B, Mistral-7B, and Flan-T5 Large and surpass previous methods on these datasets.

Fine-Tuning Large Language Models for Relation Extraction within a Retrieval-Augmented Generation Framework

This paper compares two approaches for table extraction from images: deep learning computer vision and Multimodal Large Language Models (MLLMs). Computer vision models for table extraction, such as the Table Transformer model (TATR), have enhanced the extraction of complex table structural layouts by leveraging deep learning for precise structural recognition combined with traditional Optical Character Recognition (OCR). Conversely, MLLMs, which process both text and image inputs, present a novel approach by potentially bypassing the limitations of TATR plus OCR methods altogether. Models such as GPT-4o, Phi-3 Vision, and Granite Vision 3.2 demonstrate the potential of MLLMs to analyze and interpret table images directly, offering enhanced accuracy and robust extraction capabilities. A state-of-the-art metric like Grid Table Similarity (GriTS) evaluated these methodologies, providing nuanced insights into structural and text content effectiveness. Utilizing the PubTables-1M dataset, a comprehensive and widely used benchmark in the field, this study highlights the strengths and limitations of each approach, setting the stage for future innovations in table extraction technologies. Deep learning computer vision techniques still have a slight edge when extracting table structural layout, but in terms of text cell content, MLLMs are far better.

Premium content

Downloads

Next from ACL 2025

AEHRC at BioLaySumm 2025: Leveraging T5 for Lay Summarisation of Radiology Reports

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES