{
  "id": 6448,
  "url": "https://arxiv.org/abs/2604.02091v2",
  "title": "Optimizing RAG Rerankers with LLM Feedback via Reinforcement Learning",
  "summary": "Rerankers play a pivotal role in refining retrieval results for Retrieval-Augmented Generation. However, current reranking models are typically optimized on static human annotated relevance labels in isolation, decoupled from the downstream generation process. This isolation leads to a fundamental misalignment: documents identified as topically relevant by information retrieval metrics often fail to provide the actual utility required by the LLM for precise answer generation. To bridge this gap,",
  "authors": "Yuhang Wu, Xiangqing Shen, Fanfan Wang, Cangqi Zhou, Zhen Wu, Xinyu Dai et al.",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-02T14:19:47.000Z",
  "fetched_at": "2026-07-14T16:32:28.612Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/6448",
  "original_url": "https://arxiv.org/abs/2604.02091v2",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}