← 研踪 PaperTrailarXiv:2611.02550 · v1
Re$^3$Rank: Listwise Reranking with Reasoning-Conditioned Retrieval Heads
首次提交 2026-11-02 13:24
版本更新 2026-11-02 13:24
雷达观测 2026-11-04 23:42
摘要
Attention-based rerankers score each candidate by the attention mass that retrieval heads route to it from the question, but the question, serving as a fixed query, carries limited information for locating relevant evidence. We propose Re$^3$Rank, a listwise reranker built on reasoning-conditioned retrieval heads (RRHeads). A generator first writes a reasoning trace over the question and the candidates, and a scorer then uses the trace tokens as the attention query of a few selected RRHeads, reading relevance as continuous scores. We present a straight-through estimator to propagate the ranking loss through the discrete trace, so that the generator and the scorer can be trained jointly for ranking. Averaged over 9 benchmarks, Re$^3$Rank outperforms 11 baseline rerankers: our 4B variant exceeds the strongest baseline by 3.11 NDCG@10 points on average, and even our 0.6B variant surpasses rerankers with up to 32B parameters.
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-02 13:24 · 最新观测版本
Re$^3$Rank: Listwise Reranking with Reasoning-Conditioned Retrieval Heads
Attention-based rerankers score each candidate by the attention mass that retrieval heads route to it from the question, but the question, serving as a fixed query, carries limited information for locating relevant evidence. We propose Re$^3$Rank, a listwise reranker built on reasoning-conditioned retrieval heads (RRHeads). A generator first writes a reasoning trace over the question and the candidates, and a scorer then uses the trace tokens as the attention query of a few selected RRHeads, reading relevance as continuous scores. We present a straight-through estimator to propagate the ranking loss through the discrete trace, so that the generator and the scorer can be trained jointly for ranking. Averaged over 9 benchmarks, Re$^3$Rank outperforms 11 baseline rerankers: our 4B variant exceeds the strongest baseline by 3.11 NDCG@10 points on average, and even our 0.6B variant surpasses rerankers with up to 32B parameters.
查看此版本解读 →查看此版本原文 ↗