用动态选证据的方法，让大模型更高效准确地重排长文档。
EviRerank: Adaptive Evidence Construction for Long-Document LLM Reranking
- 先用轻量模型筛选文档块，再动态分配证据权重，压缩输入长度。
- 在TREC DL'19上达0.744 nDCG@10，比现有方法更准且更省算力。
- 适合需要处理长文档的检索系统，尤其关注效率与精度平衡的场景。
解码器型大模型在长文档重排中面临推理成本高、相关性信号被稀释的问题。基于注意力分析发现，附加无关内容会削弱查询聚焦交互，本文提出EviRerank框架，通过轻量级选择器（如BM25、双编码器或交叉编码器）对文档块打分，再利用自适应证据预算（AEB）动态分配证据块，并通过摘要增强（SA）添加紧凑全局提示，在硬令牌限制下构建紧凑重排上下文，最后由解码器型大模型进行重排。在TREC DL'19、DL'22、DL'23和MLDR-zh上，EviRerank始终优于全文档重排和强基线，同时减少输入长度。RankZephyr-7B验证表明其可迁移至列表级重排。在TREC DL'19上，最高达0.744 nDCG@10和0.307 MAP，优于RankLLaMA，且使用更紧凑的证据上下文。
原文摘要 · Abstract (English)
Decoder-only LLM rerankers struggle with long documents: inference is costly and relevance signals can be diluted by irrelevant context. Motivated by a diagnostic attention analysis suggesting that appended irrelevant context can weaken query-focused interactions, we propose EviRerank, an evidence-based long-document reranking framework for decoder-only LLMs. EviRerank first scores document blocks with a lightweight selector, such as BM25, a bi-encoder, or a cross-encoder. It then constructs a compact reranking context under a hard token cap by dynamically budgeting evidence blocks with Adaptive Evidence Budgeting (AEB) and adding a compact global cue via Summary Augmentation (SA). Finally, the compact evidence context is reranked with a decoder-only LLM. Across TREC DL'19, DL'22, DL'23, and MLDR-zh, EviRerank consistently outperforms full-document LLM reranking and strong block-selection baselines while reducing input length. RankZephyr-7B validation further confirms transfer to listwise reranking. On TREC DL'19, EviRerank reaches up to 0.744 nDCG@10 and 0.307 MAP, improving over RankLLaMA while using a compact evidence context.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。