paper-with-me

Papers

GenProve: Learning to Generate Text with Fine-Grained Provenance

2026-01-08 · Jingxuan Wei, Xingyue Wang, Yanghaoyu Liao, Jie Dong, Yuchen Liu, Caijun Jia, Bihui Yu, Junnan Zhu arxiv

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.

📄 PDF Abstract BibTeX arXiv:2601.04932

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

TROVE: A Challenge for Fine-Grained Text Provenance via Source Sentence Tracing and Relationship Classification

2025-03-19 · Junnan Zhu, Min Xiao, Yining Wang, FeiFei Zhai 외

LLMs have achieved remarkable fluency and coherence in text generation, yet their widespread adoption has raised concerns about content reliability and accountability. In high-stakes domains such as healthcare, law, and …

RetrievalSentenceSentence RetrievalText Generation

Bileve: Securing Text Provenance in Large Language Models Against Spoofing with Bi-level Signature

2024-06-04 · Tong Zhou, Xuandong Zhao, Xiaolin Xu, Shaolei Ren

Text watermarks for large language models (LLMs) have been commonly used to identify the origins of machine-generated content, which is promising for assessing liability when combating deepfake or harmful content. While …

Face Swapping

What is Your Article Based On? Inferring Fine-grained Provenance

2021-08-01 · ACL 2021 5 · Yi Zhang, Zachary Ives, Dan Roth

When evaluating an article and the claims it makes, a critical reader must be able to assess where the information presented comes from, and whether the various claims are mutually consistent and support the conclusion. …

ArticlesFact CheckingSentence

Facilitating the sharing of electrophysiology data analysis results through in-depth provenance capture

2023-11-16 · Cristiano André Köhler, Danylo Ulianych, Sonja Grün, Stefan Decker 외

Scientific research demands reproducibility and transparency, particularly in data-intensive fields like electrophysiology. Electrophysiology data is typically analyzed using scripts that generate output files, including…

Transfer Learning

SMA: Who Said That? Auditing Membership Leakage in Semi-Black-box RAG Controlling

2025-08-12 · Shixuan Sun, Siyuan Liang, Ruoyu Chen, Jianjie Huang 외 arxiv

Retrieval-Augmented Generation (RAG) and its Multimodal Retrieval-Augmented Generation (MRAG) significantly improve the knowledge coverage and contextual understanding of Large Language Models (LLMs) by introducing exter…

Image Retrieval