Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning
Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning
AI总结：
提出一种基于强化学习的模态感知信用分配框架（MoCA），通过感知验证和结构化口头验证解决视觉语言模型中感知与推理的权衡问题，实现多任务性能提升。
摘要：
实现稳健的感知-推理协同是高级视觉语言模型（VLM）的核心目标。最近的进展通过架构设计或智能体工作流追求这一目标。然而，这些方法通常受限于静态文本推理，或因外部智能体复杂性的巨大计算和工程负担而变得复杂。更糟糕的是，这种大量投入并未带来成比例的性能提升，常常在感知和推理上观察到“跷跷板效应”。这促使我们从根本上重新思考真正的瓶颈。在本文中，我们认为这种权衡的根本原因是模态信用分配中的模糊性：当VLM失败时，是由于感知缺陷（“坏视力”）还是逻辑缺陷（“坏思维”）？为解决这一问题，我们引入了一个强化学习框架，通过可靠地奖励感知保真度来改善感知-推理协同。我们明确地将生成过程分解为交错的感知和推理步骤。这种解耦使得能够对感知进行有针对性的监督。关键的是，我们引入了感知验证（PV），利用“盲推理”代理独立于推理结果奖励感知保真度。此外，为了在自由形式的VL任务中扩展训练，我们提出了结构化口头验证（Structured Verbal Verification），用结构化的算法执行替代高方差的LLM评判。这些技术被整合到模态感知信用分配（MoCA）机制中，该机制将奖励路由到特定的错误源——无论是坏视力还是坏思维——使单个VLM能够在广泛的任务谱系上同时获得性能提升。
英文摘要：
Achieving robust perception-reasoning synergy is a central goal for advanced Vision-Language Models (VLMs). Recent advancements have pursued this goal via architectural designs or agentic workflows. However, these approaches are often limited by static textual reasoning or complicated by the significant compute and engineering burden of external agentic complexity. Worse, this heavy investment does not yield proportional gains, often witnessing a "seesaw effect" on perception and reasoning. This motivates a fundamental rethinking of the true bottleneck. In this paper, we argue that the root cause of this trade-off is an ambiguity in modality credit assignment: when a VLM fails, is it due to flawed perception ("bad seeing") or flawed logic ("bad thinking")? To resolve this, we introduce a reinforcement learning framework that improves perception-reasoning synergy by reliably rewarding the perception fidelity. We explicitly decompose the generation process into interleaved perception and reasoning steps. This decoupling enables targeted supervision on perception. Crucially, we introduce Perception Verification (PV), leveraging a "blindfolded reasoning" proxy to reward perceptual fidelity independently of reasoning outcomes. Furthermore, to scale training across free-form VL tasks, we propose Structured Verbal Verification, which replaces high-variance LLM judging with structured algorithmic execution. These techniques are integrated into a Modality-Aware Credit Assignment (MoCA) mechanism, which routes rewards to the specific source of error -- either bad seeing or bad thinking -- enabling a single VLM to achieve simultaneous performance gains across a wide task spectrum.
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。
