ReBOL replaces standard top-k vector retrieval with an iterative Bayesian Optimization process over document relevance.
March 24, 2026
Original Paper
ReBOL: Retrieval via Bayesian Optimization with Batched LLM Relevance Observations and Query Reformulation
arXiv · 2603.20513
The Takeaway
It treats retrieval as an acquisition problem, using LLM feedback to update a posterior over document relevance. This significantly boosts recall (up to 11.5% absolute gain) by enabling deep query-document interactions that static vector similarity misses.
From the abstract
LLM-reranking is limited by the top-k documents retrieved by vector similarity, which neither enables contextual query-document token interactions nor captures multimodal relevance distributions. While LLM query reformulation attempts to improve recall by generating improved or additional queries, it is still followed by vector similarity retrieval. We thus propose to address these top-k retrieval stage failures by introducing ReBOL, which 1) uses LLM query reformulations to initialize a multimo