感知至推理：解耦感知與推理實現細粒度視覺推理

摘要

細粒度視覺推理對視覺語言模型而言仍是一項挑戰，尤其當微小但關鍵的視覺線索隱藏在高解析度影像中時。現有方法依賴於重複裁剪或測試階段的視覺搜尋來引入局部證據，但通常未能明確區分感知與推理。本文提出感知到推理（Perceive-to-Reason, P2R）的統一框架，將細粒度視覺推理形式化為兩個階段：模型首先作為感知器定位與問題相關的證據，然後作為推理器根據標註影像與裁剪區域來回答問題。為使訓練與此分離式設計更佳對齊，我們進一步引入感知-推理交替GRPO（Perception-Reasoning Alternating GRPO, PRA-GRPO），這是一種角色感知的強化學習策略，僅利用最終答案的監督訊號，在感知聚焦與推理聚焦的更新之間交替進行。基於Qwen3-VL-Instruct-2B/4B/8B建構的P2R在模型規模上持續展現效能提升。特別地，P2R-4B在V-Star上達到93.2%，在HR-Bench-4K上達到81.9%，在HR-Bench-8K上達到80.5%，大幅超越其對應的骨幹模型。進一步實驗顯示，P2R的效益不僅限於高解析度基準，更延伸至更廣泛的多模態推理任務。這些結果表明，明確將感知與推理解耦，為細粒度視覺推理提供了一個有效的框架。

English

Fine-grained visual reasoning remains challenging for vision-language models, especially when small but critical visual cues are buried in high-resolution images. Existing approaches rely on repeated cropping or test-time visual search to introduce local evidence, but they typically do not explicitly distinguish perception from reasoning. In this paper, we propose Perceive-to-Reason (P2R), a unified framework that formulates fine-grained visual reasoning as a two-stage process: the model first localizes question-relevant evidence as a Perceiver, and then answers the question as a Reasoner based on the annotated image and cropped regions. To better align training with this decoupled formulation, we further introduce Perception-Reasoning Alternating GRPO (PRA-GRPO), a role-aware reinforcement learning strategy that alternates between perception-focused and reasoning-focused updates using only final-answer supervision. Built on top of Qwen3-VL-Instruct-2B/4B/8B, P2R consistently improves performance across model scales. In particular, P2R-4B achieves 93.2% on V-Star, 81.9% on HR-Bench-4K, and 80.5% on HR-Bench-8K, substantially outperforming its corresponding backbone. Further experiments show that the benefits of P2R extend beyond high-resolution benchmarks to broader multimodal reasoning tasks. These results suggest that explicitly decoupling perception from reasoning provides an effective framework for fine-grained visual reasoning.