让视觉推理模型自动选择最短路径，省一半以上计算量
Learning Adaptive Reasoning Paths for Efficient Visual Reasoning
- 将推理拆解为感知、逻辑和作答三步，动态选最优路径
- 在多个数据集上减少50%~90%的令牌消耗，准确率不变
- 适合需要高效推理的场景，如移动端或实时系统
视觉推理模型（VRMs）通过融合视觉感知与语言推理展现出强大的跨模态能力，但常因过度推理而生成过长的推理链。我们将其归因于推理路径冗余：许多视觉问题并不需要完整推理过程。为此，我们提出自适应视觉推理框架AVR，将视觉推理分解为视觉感知、逻辑推理和答案应用三个认知功能，并支持三种响应格式：完整格式、仅感知格式和直接作答。AVR采用FS-GRPO训练，该方法基于组相对策略优化，促使模型在保持正确性的前提下选择最高效的推理格式。在多个视觉语言基准测试中，AVR将令牌使用量降低50%~90%，尤其在感知密集型任务中表现优异。结果表明，自适应视觉推理可有效缓解VRM中的过度推理问题。代码与数据见：https://github.com/RunRiotComeOn/AVR。
原文摘要 · Abstract (English)
Visual reasoning models (VRMs) have recently shown strong cross-modal reasoning capabilities by integrating visual perception with language reasoning. However, they often suffer from overthinking, producing unnecessarily long reasoning chains for any tasks. We attribute this issue to \textbf{Reasoning Path Redundancy} in visual reasoning: many visual questions do not require the full reasoning process. To address this, we propose \textbf{AVR}, an adaptive visual reasoning framework that decomposes visual reasoning into three cognitive functions: visual perception, logical reasoning, and answer application. It further enables models to dynamically choose among three response formats: Full Format, Perception-Only Format, and Direct Answer. AVR is trained with FS-GRPO, an adaptation of Group Relative Policy Optimization that encourages the model to select the most efficient reasoning format while preserving correctness. Experiments on multiple vision-language benchmarks show that AVR reduces token usage by 50--90\% while maintaining overall accuracy, especially in perception-intensive tasks. These results demonstrate that adaptive visual reasoning can effectively mitigate overthinking in VRMs. Code and data are available at: https://github.com/RunRiotComeOn/AVR.
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。