Search papers, labs, and topics across Lattice.
This paper details the Cologne Information Retrieval group's approach to the iKAT SCAI 2026 shared task, focusing on the prediction of clarification needs in agentic conversational search systems. By leveraging advanced tools for query rewriting, retrieval, reranking, and answer generation, the study evaluates two distinct neural models for predicting when users require clarification. The findings highlight the effectiveness of these models in enhancing user interaction and information retrieval accuracy in conversational agents.
Clarification need prediction can significantly enhance user experience in conversational search systems, leading to more accurate and relevant interactions.
This paper presents the participation of the Cologne Information Retrieval group in the iKAT SCAI 2026 shared task. We use an agentic conversational search system, equipped with tools for query rewriting, retrieval and reranking, answer generation, and clarification need prediction and clarification question generation. We experiment with two different neural clarification need prediction models.
