uva-irlab-conv at SemEval-2026 Task 8: Multi-Turn RAG with Learned Sparse Retrieval and Listwise Reranking
arXiv:2606.11945v1 Announce Type: new Abstract: This report describes our participation in SemEval-2026 Task 8 on multi-turn retrieval and question answering. The task evaluates conversational systems across four domains (finance, cloud documentation, government, Wikipedia), and includes unanswerable queries where the available collection does not contain sufficient evidence to produce a complete response. We propose a multi-turn retrieval-augmented generation pipeline that combines learned sparse retrieval with LLM-based reranking and generation. Using sparse retrieval as the primary retrieva
The continuous evolution of AI models, particularly in Retrieval-Augmented Generation (RAG), drives ongoing innovation in conversational AI systems seeking to improve accuracy and relevance.
Improved RAG systems enhance the precision and reliability of AI-generated responses, crucial for enterprise applications and complex information retrieval across diverse domains.
The refined methodology for multi-turn RAG, combining sparse retrieval with LLM-based reranking, represents a more effective approach to handling complex queries and unanswerable questions.
- · AI developers
- · Conversational AI platforms
- · Enterprises with large documentation bases
- · Knowledge management systems
- · Inefficient search algorithms
- · Basic RAG implementations
More accurate and contextually relevant AI responses become standard in customer service and technical support.
Reduced operational costs for businesses through more effective automated information retrieval and problem-solving.
Enhanced trust in AI systems for critical decision-making, accelerating adoption into highly regulated industries.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL