paper-with-me

홈 › Papers

MedFact-R1: Towards Factual Medical Reasoning via Pseudo-Label Augmentation

2025-09-18 · Gengliang Li, Rongyu Chen, Bin Li, Linlin Yang, Guodong Ding arxiv

Ensuring factual consistency and reliable reasoning remains a critical challenge for medical vision-language models. We introduce MEDFACT-R1, a two-stage framework that integrates external knowledge grounding with reinforcement learning to improve the factual medical reasoning. The first stage uses pseudo-label supervised fine-tuning (SFT) to incorporate external factual expertise; while the second stage applies Group Relative Policy Optimization (GRPO) with four tailored factual reward signals to encourage self-consistent reasoning. Across three public medical QA benchmarks, MEDFACT-R1 delivers up to 22.5% absolute improvement in factual accuracy over previous state-of-the-art methods. Ablation studies highlight the necessity of pseudo-label SFT cold start and validate the contribution of each GRPO reward, underscoring the synergy between knowledge grounding and RL-driven reasoning for trustworthy medical AI. Codes are released at https://github.com/Garfieldgengliang/MEDFACT-R1.

📄 PDF Abstract BibTeX arXiv:2509.15154

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

MedFact: Benchmarking the Fact-Checking Capabilities of Large Language Models on Chinese Medical Texts

2025-09-15 · Jiayi He, Yangmin Huang, Qianyun Du, Xiangying Zhou 외 arxiv

Deploying Large Language Models (LLMs) in medical applications requires fact-checking capabilities to ensure patient safety and regulatory compliance. We introduce MedFact, a challenging Chinese medical fact-checking ben…

MedFactEval and MedAgentBrief: A Framework and Workflow for Generating and Evaluating Factual Clinical Summaries

2025-09-07 · François Grolleau, Emily Alsentzer, Timothy Keyes, Philip Chung 외 arxiv

Evaluating factual accuracy in Large Language Model (LLM)-generated clinical text is a critical barrier to adoption, as expert review is unscalable for the continuous quality assurance these systems require. We address t…

MedFact: A Large-scale Chinese Dataset for Evidence-based Medical Fact-checking of LLM Responses

2025-09-22 · Tong Chen, Zimu Wang, Yiyi Miao, Haoran Luo 외 arxiv

Medical fact-checking has become increasingly critical as more individuals seek medical information online. However, existing datasets predominantly focus on human-generated content, leaving the verification of content g…

MedFACT: Modeling Medical Feature Correlations in Patient Health Representation Learning via Feature Clustering

2022-04-21 · Xinyu Ma, Xu Chu, Yasha Wang, Hailong Yu 외

In healthcare prediction tasks, it is essential to exploit the correlations between medical features and learn better patient health representations. Existing methods try to estimate feature correlations only from data, …

ClusteringRepresentation Learning

From Generation to Collaboration: Using LLMs to Edit for Empathy in Healthcare

2026-01-22 · Man Luo, Bahareh Harandizadeh, Amara Tariq, Halim Abbas 외 arxiv

Clinical empathy is essential for patient care, but physicians need continually balance emotional warmth with factual precision under the cognitive and emotional constraints of clinical practice. This study investigates …