通过概念门控视觉蒸馏克服视觉杂乱
Overcoming Visual Clutter in Vision Language Action Models via Concept-Gated Visual Distillation
查看机构详情
- University of Technology Sydney（技术大学悉尼大学）
- Western Sydney University（西悉尼大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本研究提出概念门控视觉蒸馏方法，通过视觉蒸馏技术解决杂乱环境中视觉语言动作模型的精度-推理间隙问题，显著提升机器人操作的成功率。
中文摘要 AI 辅助
视觉语言动作（VLA）模型在零样本泛化方面表现出色，但在杂乱环境中经常出现'精度-推理间隙'的问题。这种失败是由背景引起的特征稀释驱动的，其中高频语义噪声会腐蚀精确操作所需的几何定位。为弥合这一差距，我们提出了概念门控视觉蒸馏（CGVD），这是一种无需训练、模型无关的推理框架，能够稳定VLA策略。CGVD通过将指令解析为安全和干扰集，利用两层目标细化过程——结合交叉验证和空间歧义消除——来显式惩罚假阳性并隔离真正的操作目标。然后通过基于傅里叶的修复生成一个干净的观察，该观察主动抑制语义干扰物，同时保持关键空间几何和视觉本体感觉。在高度杂乱的操作任务中进行的广泛评估表明，CGVD防止了性能崩溃。在密集语义干扰物的环境中，我们的方法显著优于最先进的基线，达到77.5%的成功率，而基线仅为43.0%。通过强制严格属性遵守，CGVD将推理时的视觉蒸馏确立为在杂乱环境中实现鲁棒机器人操作的关键前提。
英文摘要
Vision-Language-Action (VLA) models demonstrate impressive zero-shot generalization but frequently suffer from a "Precision-Reasoning Gap" in cluttered environments. This failure is driven by background-induced feature dilution, where high-frequency semantic noise corrupts the geometric grounding required for precise manipulation. To bridge this gap, we propose Concept-Gated Visual Distillation (CGVD), a training-free, model-agnostic inference framework that stabilizes VLA policies. CGVD operates by parsing instructions into safe and distractor sets, utilizing a two-layer target refinement process--combining cross-validation and spatial disambiguation--to explicitly penalize false positives and isolate genuine manipulation targets. We then process the scene via Fourier-based inpainting, generating a clean observation that actively suppresses semantic distractors while preserving critical spatial geometry and visual proprioception. Extensive evaluations in highly cluttered manipulation tasks demonstrate that CGVD prevents performance collapse. In environments with dense semantic distractors, our method significantly outperforms state-of-the-art baselines, achieving a 77.5% success rate compared to the baseline's 43.0%. By enforcing strict attribute adherence, CGVD establishes inference-time visual distillation as a critical prerequisite for robust robotic manipulation in the clutter.
