3D-VCD：通过视觉对比解码缓解3D-LLM具身代理中的幻觉
3D-VCD: Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive Decoding
AI总结：
本文提出3D-VCD，一种用于缓解3D具身代理幻觉的视觉对比解码框架，通过构造扭曲的3D场景图来提升 grounded 推理能力，实验表明其在3D-POPE和HEAL基准上有效。
摘要：
大型多模态模型日益被用作在3D环境中操作的具身代理的推理核心，但它们仍容易产生幻觉，导致不安全且不基于现实的决策。现有推理时幻觉缓解方法主要针对2D视觉-语言设置，无法转移到具身3D推理中，其中失败源于物体存在、空间布局和几何基础，而非像素级不一致。我们引入3D-VCD，首个推理时视觉对比解码框架用于缓解3D具身代理的幻觉。3D-VCD通过将语义和几何扰动应用于以物体为中心的表示，如类别替换和坐标或范围破坏，构建扭曲的3D场景图。通过对比原始和扭曲的3D上下文下的预测，我们的方法抑制了对基于场景证据不敏感的标记，这些标记因此可能由语言先验驱动。我们评估了3D-VCD在3D-POPE和HEAL基准上的表现，显示其在不重新训练的情况下持续提升 grounded 推理，确立了基于结构化3D表示的推理时对比解码作为更可靠具身智能的有效且实用路径。
英文摘要：
Large multimodal models are increasingly used as the reasoning core of embodied agents operating in 3D environments, yet they remain prone to hallucinations that can produce unsafe and ungrounded decisions. Existing inference-time hallucination mitigation methods largely target 2D vision-language settings and do not transfer to embodied 3D reasoning, where failures arise from object presence, spatial layout, and geometric grounding rather than pixel-level inconsistencies. We introduce 3D-VCD, the first inference-time visual contrastive decoding framework for hallucination mitigation in 3D embodied agents. 3D-VCD constructs a distorted 3D scene graph by applying semantic and geometric perturbations to object-centric representations, such as category substitutions and coordinate or extent corruption. By contrasting predictions under the original and distorted 3D contexts, our method suppresses tokens that are insensitive to grounded scene evidence and are therefore likely driven by language priors. We evaluate 3D-VCD on the 3D-POPE and HEAL benchmarks and show that it consistently improves grounded reasoning without any retraining, establishing inference-time contrastive decoding over structured 3D representations as an effective and practical route to more reliable embodied intelligence.
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。
