Call for Papers
GenBench is an EMNLP-co-located workshop on evaluating and benchmarking generalisation in NLP models. It invites original contributions on generalisation in NLP, benchmarking, evaluation methodology, distribution shift, compositional generalisation, LLM evaluation, collaborative benchmarking. Submissions may take the form of full research papers, extended abstracts and poster presentations reporting theoretical, computational and experimental work, as well as industrial case studies and applications. The meeting brings together researchers, engineers, industry practitioners and students to exchange recent results and discuss emerging directions in the field. Accepted papers are published in the conference proceedings (Association for Computational Linguistics).
