用自对齐学习提升RAG证据提取质量，更准更短更可信
SEER: Self-Aligned Evidence Extraction for Retrieval-Augmented Generation
- 通过自对齐学习训练模型自动提取证据，替代人工规则
- 证据长度减少9.25倍，且忠实度、帮助性、简洁性全面提升
- 适合需要高效精准生成的对话与问答系统使用
检索增强生成（RAG）近期研究探索从检索到的文本中提取证据以降低计算开销并提升性能，但仍面临挑战。现有方法高度依赖基于启发式的增强，存在三方面问题：(1) 因手工设计的上下文过滤导致泛化能力差；(2) 基于规则的上下文切分造成语义缺失；(3) 句子级筛选学习导致长度分布失衡。为此，我们提出模型驱动的证据提取学习框架SEER，通过自对齐学习优化基础模型，使其具备理想的证据提取特性。大量实验表明，该方法显著提升最终RAG性能，增强提取证据的忠实度、帮助性和简洁性，并将证据长度压缩至原来的1/9.25。代码将开源于https://github.com/HITsz-TMG/SEER。
原文摘要 · Abstract (English)
Recent studies in Retrieval-Augmented Generation (RAG) have investigated extracting evidence from retrieved passages to reduce computational costs and enhance the final RAG performance, yet it remains challenging. Existing methods heavily rely on heuristic-based augmentation, encountering several issues: (1) Poor generalization due to hand-crafted context filtering; (2) Semantics deficiency due to rule-based context chunking; (3) Skewed length due to sentence-wise filter learning. To address these issues, we propose a model-based evidence extraction learning framework, SEER, optimizing a vanilla model as an evidence extractor with desired properties through self-aligned learning. Extensive experiments show that our method largely improves the final RAG performance, enhances the faithfulness, helpfulness, and conciseness of the extracted evidence, and reduces the evidence length by 9.25 times. The code will be available at https://github.com/HITsz-TMG/SEER.
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。