BioNLP Workshop
SIGBIOMED | BioNLP 2024 | BioNLP 2023
BIONLP 2025 and Shared Tasks @ ACL 2025
The 24th BioNLP workshop associated with the ACL SIGBIOMED special interest group is co-located with ACL 2025
IMPORTANT DATES (Tentative)
- Paper submission deadline: March 20, 2025
- Notification of acceptance: April 20, 2025
- Camera-ready paper due: May 20, 2025 -- No extensions due to ACL publication deadline.
- Pre-recorded video due (hard deadline): July 7, 2025
- Workshop: July 31st OR August 1st 2025
SUBMISSION INSTRUCTIONS
Two types of submissions are invited: full (long) papers (8 pages) and short papers (4 pages).
Submission site for the workshop https://softconf.com/acl2025/BioNLP2025 Submission site for Shared Tasks https://softconf.com/acl2025/BioNLP2025-ST
Please follow these formatting guidelines: https://github.com/acl-org/acl-style-files Please note that the review process is double-blind.
Final versions of accepted papers will be given one additional page of content (up to 9 pages for long papers, up to 5 pages for short papers) to address reviewers’ comments.
Submissions from ACL rolling review
We will consider ACL rolling review submissions with all reviews and scores. If you are interested in submitting your work for consideration, please contact ddemner at gmail.
WORKSHOP OVERVIEW AND SCOPE
The BioNLP workshop, associated with the ACL SIGBIOMED special interest group, is an established primary venue for presenting research in language processing and language understanding for the biological and medical domains. The workshop has been running every year since 2002 and continues getting stronger. Many other emerging biomedical and clinical language processing workshops can afford to be more specialized because BioNLP truly encompasses the breadth of the domain and brings together researchers in biomedical and clinical NLP from all over the world.
BioNLP 2025 will be particularly interested in evaluation frameworks and metrics that reflect the needs of health-related use cases and provide a good estimate of reliability of the proposed solutions. BioNLP 2025 continues to focus on transparency of tgenerative approaches and factuality of the generated text. Language processing that supports DEIA (Diversity, Equity, Inclusion and Accessibility) continues to be of utmost importance. The work on detection and mitigation of bias and misinformation continues to be of interest. Research in languages other than English, particularly, under-represented languages, and health disparities are always of interest to BioNLP. Other active areas of research include, but are not limited to:
- Extraction of complex relations and events;
- Discourse analysis; Anaphora \& coreference resolution;
- Text mining \& Literature based discovery;
- Question Answering; Summarization; Text simplification;
- Resources and strategies for system testing and evaluation;
- Synthetic data generation \& data augmentation;
- Translating NLP research into practice: tangible explainable results of biomedical language processing applications.
SHARED TASKS
SMAFIRA
The SMAFIRA project supports finding alternative methods to animal experiments. The organizers have released SMAFIRA Web tool that allows researchers to perform searches for methods alternative to animal experiments. The input to the tool is a PubMed identifier (PMID) of a publication that represents the animal experiment for which one wants to find an alternative method. The tool retrieves up to 200 similar articles available in PubMed, and presents these as a list of results. The task is to validate and annotate the top 10 similar articles, either automatically, with any system of the participants choice, or manually using the SMAFIRA tool. See details at https://smafira-bf3r.github.io/smafira-st/
ClinIQLink 2025 - LLM Lie Detector Test
The LLM Lie Detector Test aims to evaluate the effectiveness of generative models in producing factually accurate information, with a benchmark dataset specifically curated to align with the knowledge level of a General Practitioner (GP) Medical Doctor. Participants will submit model ouptputs to be assessed using a structured set of atomic question-answer pairs (factoid, true/false and list questions), which focus on retrieving precise, factually correct information. The test will evaluate internal model knowledge retrieval. See details at https://brandonio-c.github.io/ClinIQLink-2025/
ArchEHR-QA 2025: Grounded Electronic Health Record Question Answering
The participants will automatically generate answers to patients’ health-related questions that are grounded in the evidence from patients’ clinical notes. The dataset will consist of hand-curated realistic patient questions (submitted through a patient portal) and their corresponding clinician-rewritten versions (crafted to assist in formulating their responses). The task is to construct coherent answers or responses to input questions that must use and be grounded in the provided clinical note excerpts. See details at https://archehr-qa.github.io/
BioLaySumm 2025
This is the 3nd iteration of BioLaySumm, following the success of the 2nd edition of the task at BioNLP 2024 which attracted 200 plus submissions across 53 different teams and the 1st edition of the task at BioNLP 2023 which attracted 56 submissions across 20 different teams. This edition builds on last year’s task by introducing a new task: radiology report generation with layman’s terms, extending the shared task to a new domain and multi-modality. See detail at https://biolaysumm.org/