让视觉语言模型像人一样一步步推理，每步都有视觉证据支持。
H-GRPO: Permutation-Invariant Reinforcement Learning for Grounded Visual Reasoning
- 将复杂问题拆解为一系列带视觉定位的子问题，强制逐步推理。
- 通过显式验证每个逻辑步骤的视觉依据，提升答案可靠性。
- 适合需要可解释性推理的场景，如医疗图像分析或自动驾驶决策。
视觉语言模型（VLMs）在基准测试中表现优异，但常被视为黑箱，易产生幻觉或依赖表面捷径。本文提出一种新框架H-GRPO，通过分解性证据定位增强性能与可解释性。不同于整体推理方式，该方法迫使模型将全局查询分解为一系列原子级子问题，每个子问题需提供明确的子答案及对应的局部视觉证据框。通过将中间逻辑步骤（如识别容器、分析液体属性、评估环境背景）与特定视觉区域关联，构建出类人类推理的结构化路径。最终答案由此前经验证的视觉事实逻辑推导得出，而非统计猜测。
原文摘要 · Abstract (English)
Vision-Language Models (VLMs) often achieve high performance on benchmarks while remaining "black boxes", yet they remain prone to hallucination or rely on superficial shortcuts. In this work, we propose a framework designed to enhance both performance and interpretability through De-compositional Evidence Grounding. Unlike monolithic inference approaches, our approach forces the model to decompose a global query into a sequence of atomic sub-questions, each requiring an explicit sub-answer and critically a localized evidence bounding box. By grounding intermediate logical steps (e.g. identifying a container, analyzing liquid properties, and assessing environmental context) in specific visual regions, we construct a structured reasoning path that mirrors human-like deduction. This allows the final answer to emerge as a logical consequence of verified visual facts rather than a statistical guess.
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。