UniReason-Med: 用于医学VQA中二维到三维迁移的共享基础推理接口
UniReason-Med: A Shared Grounded Reasoning Interface for 2D-to-3D Transfer in Medical VQA
AI总结：
提出UniReason-Med框架，通过共享基础推理接口从2D医学图像向3D医学VQA迁移推理能力，结合监督微调和强化学习，显著提升3D推理性能。
摘要：
我们研究了当两种输入类型通过共同的推理接口对齐时，来自丰富2D医学图像的基础推理监督是否能够改善3D医学VQA。我们引入了UniReason-Med，一个单一检查点框架，在推理时处理2D图像或切片序列化的3D体积，通过共享框语法、区域标记注入和共同的基础推理策略生成交错文本推理和局部视觉证据。为了训练这个接口，我们构建了UniMed-CoT，一个包含220K指令微调数据集，具有交错的文本推理和基础视觉证据，包括170K 2D和50K 3D样本。通过监督微调后接结果级强化学习，UniReason-Med学会生成基础推理轨迹，而在强化学习期间无需基于IoU/Dice的定位奖励。数据混合和组件消融实验表明，联合2D+3D基础监督显著改善了仅3D训练的3D推理，而基础化和区域标记注入对2D和3D任务都有持续益处。这些结果表明，共享的基础推理接口可以将推理结构从2D图像迁移到切片序列化的体积医学理解。代码和数据公开在https://this URL。
英文摘要：
We study whether grounded reasoning supervision from abundant 2D medical images can improve 3D medical VQA when both input types are aligned through a common reasoning interface. We introduce UniReason-Med, a single-checkpoint framework that processes either a 2D image or a slice-serialized 3D volume at inference time, generating interleaved textual reasoning and localized visual evidence through shared box syntax, region-token injection, and a common grounded reasoning policy. To train this interface, we construct UniMed-CoT, a 220K instruction-tuning dataset with interleaved textual reasoning and grounded visual evidence, including 170K 2D and 50K 3D samples. Through supervised fine-tuning followed by outcome-level reinforcement learning, UniReason-Med learns to generate grounded reasoning traces without IoU/Dice-based localization rewards during RL. Data-mixture and component ablations show that joint 2D+3D grounded supervision substantially improves 3D reasoning over 3D-only training, while grounding and region-token injection consistently benefit both 2D and 3D tasks. These results suggest that a shared grounded reasoning interface can transfer reasoning structure from 2D images to slice-serialized volumetric medical understanding. The code and data are publicly available at https://github.com/IQuestLab/unireason-med.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
