VisRef：思考时视觉重聚焦提升多模态大推理模型的测试时扩展
VisRef: Visual Refocusing while Thinking Improves Test-Time Scaling in Multi-Modal Large Reasoning Models
提出一种无需额外强化学习微调的测试时扩展框架，通过选择与推理上下文语义相关且全局多样的视觉令牌核心集并重新注入，缓解多模态大模型在长文本推理中对视觉信息的注意力衰减，可复用于视觉问答、视觉推理等任务。
原文依据 re-injecting a coreset of visual tokens that are semantically relevant to the reasoning context yet diverse and globally representative of the image
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论大型推理模型的进展表明，通过扩展推理时的思考来增加测试时计算量，可在复杂推理任务上取得强劲性能。然而，近期研究观察到，在依赖视觉的任务中，推理时延长的文本推理往往会导致性能下降，因为模型逐渐对视觉令牌失去注意力，越来越依赖纯文本先验。为解决此问题，先前工作采用基于强化学习（RL）的微调来路由视觉令牌，或在推理过程中引入重聚焦机制。这些方法虽然有效，但计算成本高昂，需要大规模数据生成和策略优化。为了在不额外进行RL微调的情况下利用推理时计算的优势，我们提出VisRef，一个视觉接地的测试时扩展框架。我们的核心思想是主动引导推理过程：重新注入一个视觉令牌核心集，这些令牌与推理上下文语义相关，同时具有多样性且能全局代表图像，从而实现更接地的多模态推理。在三个视觉推理基准上使用最优多模态大推理模型进行的实验表明，在固定的推理时计算预算下，VisRef始终优于现有的测试时扩展方法，提升幅度最高达6.4%。
英文原文摘要 1342 字符
Advances in large reasoning models have shown strong performance on complex reasoning tasks by scaling test-time compute through extended inference-time thinking. However, recent studies observe that in vision-dependent tasks, extended textual reasoning at inference time can often degrade performance as models progressively lose attention to visual tokens, increasingly relying on textual priors alone. To address this, prior works used reinforcement learning (RL)-based fine-tuning to route visual tokens or employ refocusing mechanisms during reasoning. While effective, these methods are computationally expensive, requiring large-scale data generation and policy optimization. To leverage the benefits of inference-time compute without additional RL fine-tuning, we propose VisRef, a visually grounded test-time scaling framework. Our key idea is to actively guide the reasoning process through re-injecting a coreset of visual tokens that are semantically relevant to the reasoning context yet diverse and globally representative of the image for more grounded multi-modal reasoning. Experiments on three visual reasoning benchmarks with state-of-the-art multi-modal large reasoning models demonstrate that under fixed inference-time compute budgets, VisRef consistently outperforms existing test-time scaling approaches by up to 6.4%.