GenProve: Learning to Generate Text with Fine-Grained Provenance

• URL: http://arxiv.org/abs/2601.04932v1
• Date: Thu, 08 Jan 2026 13:30:30 GMT
• Title: GenProve: Learning to Generate Text with Fine-Grained Provenance
• Abstract summary: We introduce Generation-time Fine-grained Provenance, a task where models must generate fluent answers while simultaneously producing structured, sentence-level provenance triples.<n>By optimizing a composite reward for answer fidelity and provenance correctness, GenProve significantly outperforms 14 strong LLMs in joint evaluation.
• Score: 14.66501508318216
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• GRACE: Graph-Grounded Reflective Agent Copilot Engine for Expert-in-the-Loop Knowledge Expansion [2.729876880765472]
  Large language models frequently generate plausible but ungrounded claims.<n>We introduce GRACE, a framework that deconstructs responses into atomic claims and grounds them against trusted knowledge priors.<n>We evaluate GRACE across multiple language models and on datasets spanning both general and domain-specific knowledge.
  arXiv  Detail & Related papers  (2026-09-03T20:03:20Z)
• Reliability-Prioritized Fine-Grained Generation in Multimodal Large [50.26126844186488]
  We show that generating fine-grained descriptions is more error-prone than coarse-grained generation.<n>We propose a hierarchy-aware evaluation algorithm, which assesses both whether model predictions are visually correct and how specific the correct predictions are.<n>Experiments on textscGranFact show that our method improves fine-grained generation while preserving reliability.
  arXiv  Detail & Related papers  (2026-06-28T19:27:59Z)
• Guaranteeing Knowledge Integration with Joint Decoding for Retrieval-Augmented Generation [45.13215113490545]
  GuarantRAG is a framework that explicitly decouples reasoning from evidence integration.<n> Experiments on five QA benchmarks demonstrate that GuarantRAG improves accuracy by up to 12.1%.
  arXiv  Detail & Related papers  (2026-04-09T09:52:21Z)
• Are Finer Citations Always Better? Rethinking Granularity for Attributed Generation [75.27453112832093]
  We show that enforcing fine-grained citations degrades attribution quality by 16-276% compared to the best-performing granularity.<n>Our analysis suggests that fine-grained (sentence-level) citations disrupt necessary semantic dependencies for attributing evidence to answer claims.<n>Strikingly, citation-optimal granularity leads to substantial gains in attribution quality while preserving or even improving answer correctness.
  arXiv  Detail & Related papers  (2026-04-01T22:16:23Z)
• SimGR: Escaping the Pitfalls of Generative Decoding in LLM-based Recommendation [68.00727783181289]
  A core objective in recommender systems is to accurately model the distribution of user preferences over items to enable personalized recommendations.<n>We observe that existing methods inevitably introduce systematic bias when estimating item-level preference distributions.<n>We propose textbfSimply textbfGenerative textbfRecommendation (textbfSimGR), a framework that directly models item-level preference distributions in a shared latent space.
  arXiv  Detail & Related papers  (2026-02-08T07:26:52Z)
• Coupled Variational Reinforcement Learning for Language Model General Reasoning [83.82392089177841]
  We propose textitbCoupled bVari bReinforcement bLearning (CoVRL) to bridge variational inference and reinforcement learning.<n>CoVRL improves performance by 12.4% over the base model and achieves an additional 2.3% improvement over strong state-of-the-art verifier-free RL baselines.
  arXiv  Detail & Related papers  (2025-12-14T07:03:51Z)
• RouteRAG: Efficient Retrieval-Augmented Generation from Text and Graph via Reinforcement Learning [69.87510139069218]
  Retrieval-Augmented Generation (RAG) integrates non-parametric knowledge into Large Language Models (LLMs)<n>Recent progress has advanced text-based RAG to multi-turn reasoning through Reinforcement Learning (RL)<n>We introduce model, an RL-based framework that enables LLMs to perform multi-turn and adaptive graph-text hybrid RAG.
  arXiv  Detail & Related papers  (2025-12-10T10:05:31Z)
• Decomposition-Enhanced Training for Post-Hoc Attributions In Language Models [64.49342399229529]
  We argue that post-hoc attribution can be reframed as a reasoning problem, where answers are decomposed into constituent units, each tied to specific context.<n>We introduce DecompTune, a post-training method that teaches models to produce answer decompositions as intermediate reasoning steps.<n>Across extensive experiments and ablations, DecompTune substantially improves attribution quality, outperforming prior methods and matching or exceeding state-of-the-art frontier models.
  arXiv  Detail & Related papers  (2025-10-29T17:58:59Z)
• VeriCite: Towards Reliable Citations in Retrieval-Augmented Generation via Rigorous Verification [107.75781898355562]
  We introduce a novel framework, called VeriCite, designed to rigorously validate supporting evidence and enhance answer attribution.<n>We conduct experiments across five open-source LLMs and four datasets, demonstrating that VeriCite can significantly improve citation quality while maintaining the correctness of the answers.
  arXiv  Detail & Related papers  (2025-10-13T13:38:54Z)
• Writing-Zero: Bridge the Gap Between Non-verifiable Tasks and Verifiable Rewards [11.149294285483782]
  We propose a unified RLVR-based training paradigm that bridges the gap between non-verifiable tasks and verifiable rewards.<n>We introduce a writing-principle-based pairwise Generative Reward Model (GenRM) and a novel Bootstrapped Relative Policy Optimization (BRPO) algorithm.<n>Our approach empowers LLMs to develop robust writing capabilities without supervised fine-tuning.
  arXiv  Detail & Related papers  (2025-05-30T14:34:57Z)
• Ground Every Sentence: Improving Retrieval-Augmented LLMs with Interleaved Reference-Claim Generation [51.8188846284153]
  Attributed Text Generation (ATG) is proposed to enhance credibility and verifiability in RAG systems.<n>This paper proposes ReClaim, a fine-grained ATG method that alternates the generation of references and answers step by step.<n>With extensive experiments, we verify the effectiveness of ReClaim in extensive settings, achieving a citation accuracy rate of 90%.
  arXiv  Detail & Related papers  (2024-07-01T20:47:47Z)
• Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation [8.975024781390077]
  We present MIRAGE --Model Internals-based RAG Explanations -- a plug-and-play approach using model internals for faithful answer attribution in question answering applications. We evaluate our proposed approach on a multilingual QA dataset, finding high agreement with human answer attribution.
  arXiv  Detail & Related papers  (2024-06-19T16:10:26Z)
• Training Language Models to Generate Text with Citations via Fine-grained Rewards [19.176465185343417]
  Large Language Models (LLMs) are prone to hallucination, and their responses often lack credibility due to missing references to reliable sources. We propose an effective training framework using fine-grained rewards to teach LLMs to generate highly supportive and relevant citations. On LLaMA-2-7B, the incorporation of fine-grained rewards achieves the best performance among the baselines, even surpassing that of GPT-3.5-turbo.
  arXiv  Detail & Related papers  (2024-02-06T19:00:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.