arXiv:2606.16783cs.CVcs.AI2026-06被引 1

用生成的图像作为视觉推理中间步骤,提升多模态模型可解释性。

Gen-VCoT: Generative Visual Chain-of-Thought Reasoning via Diffusion-Based RGB Intermediate Representations

论文配图:Gen-VCoT: Generative Visual Chain-of-Thought Reasoning via Diffusion-Based RGB Intermediate Representations
图 1 · 摘自论文原文
  • 通过分割、深度图和视觉语言模型生成三阶段图像中间表示。
  • 空间和深度问题准确率分别提升25%和50%,但简单事实题表现下降。
  • 适合需要可解释推理的复杂视觉任务,如三维理解与场景分析。

多模态大模型在视觉推理上表现优异,但依赖文本链式思考(CoT),缺乏可解释的视觉中间表示。现有方法使用不透明的令牌或外部工具,难以捕捉关键特性。本文提出Gen-VCoT框架,利用专家视觉模型生成RGB图像作为推理中间表示,包含三个阶段:视觉定位(SAM分割)、几何推理(Marigold深度图)和语义推理(Qwen2-VL整合)。自适应路由机制选择推理深度。评估显示,该方法在空间(提升25%)和深度(提升50%)类问题上表现更优,但在简单事实问答上表现不如文本CoT(CLEVR数据集上91.2% vs 62.5%),表明最优表示具有任务依赖性。Gen-VCoT建立了一种新的可解释多模态推理范式。

原文摘要 · Abstract (English)

Multimodal large language models (MLLMs) excel at visual reasoning but rely on text-based chain-of-thought (CoT), lacking interpretable visual intermediates. Existing methods use opaque tokens or external tools, missing key properties. We propose Gen-VCoT, a framework using expert vision models to generate RGB images as reasoning intermediates. It has three stages: visual grounding (SAM segmentation), geometric reasoning (Marigold depth maps), and semantic reasoning (Qwen2-VL integration). An adaptive router selects reasoning depth. Evaluations show Gen-VCoT improves spatial (25% better) and depth (50% better) questions, but may hurt simple factual queries. Text CoT outperforms visual intermediates on CLEVR (91.2% vs 62.5%), showing task-dependent optimal representations. Gen-VCoT establishes a new paradigm for interpretable multimodal reasoning.

视觉推理扩散模型可解释性多模态

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。