RAG效果不靠堆信息，关键在模型能否用得上。
Metadata, Structure, or Strategy? A Decomposition of RAG Context Enrichment
- 拆解元数据、结构、策略三因素，控制实验验证
- 多数增强反而降低准确率，最高差19F1点
- 适合谁看：做RAG优化或模型适配的研究者
检索增强生成（RAG）系统常通过附加质量元数据、结构化文档、多跳检索策略来丰富上下文，假设更丰富的上下文带来更好回答。然而现有评估无法分离这三个因素的影响。本研究在六个基准上，使用来自三个模型家族的四款模型，对五种增强级别进行超过24,000次响应评估，发现该假设不成立：多数增强反而降低准确率。模型虽能正确利用置信度分数，但答案质量下降，暴露了使用与准确性的差距，这是此前研究未测量的。决定答案质量的关键不是上下文携带的元数据量，而是模型能否有效利用这些信息。当元数据和检索策略与模型能力对齐时，小型模型比前沿模型高出19 F1分。研究提出一个可处理性层级，仅基于预训练特性即可预测模型能有效使用何种元数据，将RAG设计重新定义为模型-上下文对齐问题，而非元数据积累。
原文摘要 · Abstract (English)
Retrieval-augmented generation (RAG) systems increasingly enrich retrieved passages by attaching quality metadata, structuring them into explicit records, and adopting multi-hop retrieval strategies that accumulate evidence across steps. These changes assume that richer context yields better answers, yet existing evaluations cannot test this because they vary all three factors at once. We isolate each factor in a controlled experiment across six benchmarks, four models from three families, and five enrichment levels, totaling over 24,000 evaluated responses. The assumption does not hold. Most enrichment reduces accuracy. Models prompted to use confidence scores comply correctly yet produce worse answers, a gap between utilization and accuracy that no prior work has measured. What determines answer quality is not how much metadata the context carries but whether the model can act on it for the given task. When metadata and retrieval strategy are aligned with model capabilities, a smaller model outperforms a frontier model by 19 F1 points. These findings motivate a processability hierarchy that predicts, from pre-training properties alone, which metadata a model can productively use, reframing RAG design as a question of model-context alignment rather than metadata accumulation.
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。