让大模型生成时附带精准来源，区分引用与推理。
GenProve: Learning to Generate Text with Fine-Grained Provenance
- 生成答案时同步输出句子级来源三元组，精细标注引用、压缩、推断。
- 在14个主流模型中，联合评估表现最优，证明来源准确率显著提升。
- 揭示模型在推理类来源上仍薄弱，适合关注可验证推理的研究者。
大语言模型常产生幻觉，尽管添加引用是常见应对方式，但用户难以验证引用内容是否真正支持生成结论。现有方法多为粗粒度，无法区分直接引用与复杂推理。本文提出生成阶段细粒度溯源任务，要求模型在生成流畅回答的同时，输出结构化的句子级溯源三元组。为此，我们构建了ReFInE数据集，包含专家标注的引用、压缩、推断三类细粒度来源。基于此，我们提出GenProve框架，结合监督微调（SFT）与组相对策略优化（GRPO），通过联合优化答案忠实度与溯源正确性，使模型在14个强基线模型的联合评估中表现更优。关键发现：模型在表面引用上表现良好，但在推理类溯源上存在明显能力缺口，表明可验证推理仍是独立挑战。
原文摘要 · Abstract (English)
Large language models (LLM) often hallucinate, and while adding citations is a common solution, it is frequently insufficient for accountability as users struggle to verify how a cited source supports a generated claim. Existing methods are typically coarse-grained and fail to distinguish between direct quotes and complex reasoning. In this paper, we introduce Generation-time Fine-grained Provenance, a task where models must generate fluent answers while simultaneously producing structured, sentence-level provenance triples. To enable this, we present ReFInE (Relation-aware Fine-grained Interpretability & Evidence), a dataset featuring expert verified annotations that distinguish between Quotation, Compression, and Inference. Building on ReFInE, we propose GenProve, a framework that combines Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). By optimizing a composite reward for answer fidelity and provenance correctness, GenProve significantly outperforms 14 strong LLMs in joint evaluation. Crucially, our analysis uncovers a reasoning gap where models excel at surface-level quotation but struggle significantly with inference-based provenance, suggesting that verifiable reasoning remains a frontier challenge distinct from surface-level citation.
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。