uva-irlab-conv at SemEval-2026 Task 8: Multi-Turn RAG with Learned Sparse Retrieval and Listwise Reranking
Published in SemEval 2026, 2026
This paper presents a multi-turn retrieval-augmented generation pipeline combining learned sparse retrieval with conversational query rewriting, listwise reranking, and answer generation conditioned on the full conversation history.
Recommended citation: Lupart, S., Mekonnen, K., Abbasiantaeb, Z., and Aliannejadi, M. (2026). "uva-irlab-conv at SemEval-2026 Task 8: Multi-Turn RAG with Learned Sparse Retrieval and Listwise Reranking." Proceedings of the 20th International Workshop on Semantic Evaluation, 2972–2980.
Download Paper