📄 Proceedings of the Fourth Workshop on Bridging Human-Computer Interaction and Natural Language Processing (HCI+NLP)

Summary

These are the proceedings of the fourth HCI+NLP workshop. The event took place on 9 November 2025 as part of EMNLP 2025. Its goal is to bring two communities closer together: natural language processing (NLP) and human-computer interaction (HCI). For NLP researchers, the workshop shows how HCI methods help evaluate and improve language technologies. These methods include studying user behavior, human-centered design and bias auditing. The HCI community, in turn, learns how large language models are changing interfaces and research practice. The organizers single out problems that have become more acute with the spread of LLMs: evaluation in real-world conditions, bias and fairness, and model interpretability. The program included 38 papers, 25 of which are published in the volume. The topics are broad: trust and control in interaction with LLMs, verification of biomedical claims, speech recognition for elderly patients, mobile GUI agents, teen safety on TikTok, cultural sensitivity of dialogue systems, and readability assessment. For Mova, the proceedings are valuable as an overview of human-centered evaluation approaches that can be applied to Ukrainian language tools.

Key points

• This is the fourth edition of the workshop. Previous editions were held at EACL 2021, NAACL 2022 and NAACL 2024. The organizing committee consists of nine people from Microsoft Research, Google Research, Adobe Research, CMU, Stanford, Johns Hopkins and other institutions.
• 38 papers were accepted: 25 archival (included in the volume) and 13 non-archival (presentations at the workshop only).
• The preface names two invited speakers: Anjalie Field (Johns Hopkins) and Zhicong Lu (George Mason). The committee list also includes Heloisa Candello (IBM Research).
• The systems and benchmarks presented include TripleCheck (post-hoc verification of biomedical claims in AI responses), MobileA3gent (training mobile GUI agents on decentralized data), MEETING DELEGATE (a benchmark for LLMs that take part in meetings on a person's behalf) and Re:Member (generating emotional questions from personal memories). There is also a benchmark of culturally aware conversations for LLMs.
• Publisher: ACL, ISBN 979-8-89176-353-1.