ProGAL-VLA: 通过前瞻性推理实现视觉-语言-动作模型中的 grounded 对齐
ProGAL-VLA: Grounded Alignment through Prospective Reasoning in Vision-Language-Action Models
AI总结：
ProGAL-VLA 通过构建 3D 实体中心图、使用慢速规划器生成符号子目标，并通过 Grounding Alignment Contrastive 损失对齐实体，提升机器人在扰动下的鲁棒性，减少语言忽略并提高实体检索性能。
摘要：
视觉语言动作（VLA）模型能够使通用机器人代理，但通常表现出语言无知，依赖视觉捷径且对指令变化不敏感。我们提出了前瞻性接地和对齐 VLA（ProGAL-VLA），它构建了一个 3D 实体中心图（GSM），使用慢速规划器生成符号子目标，并通过接地对齐对比损失（GAC 损失）将这些子目标与接地实体对齐。所有动作都基于经过验证的目标嵌入 $g_t$，其注意力熵提供内在模糊性信号。在 LIBERO-Plus 上，ProGAL-VLA 在机器人扰动下的鲁棒性从 30.3% 提高到 71.5%，减少了 3-4 倍的语言无知，将实体检索从 0.41 提高到 0.71 Recall@1。在自定义模糊性基准上，它达到 AUROC 0.81（vs. 0.52）、AUPR 0.79，并在不损害无歧义成功的情况下将模糊输入的澄清率从 0.09 提高到 0.81。验证瓶颈增加了语言-动作之间的互信息，GAC 损失对实体施加了 InfoNCE 绑定，注意力熵产生了校准的预测选择，表明显式验证的接地是实现指令敏感、模糊意识代理的有效路径。
英文摘要：
Vision language action (VLA) models enable generalist robotic agents but often exhibit language ignorance, relying on visual shortcuts and remaining insensitive to instruction changes. We present Prospective Grounding and Alignment VLA (ProGAL-VLA), which constructs a 3D entity-centric graph (GSM), uses a slow planner to produce symbolic sub-goals, and aligns them with grounded entities via a Grounding Alignment Contrastive (GAC) loss. All actions are conditioned on a verified goal embedding $g_t$, whose attention entropy provides an intrinsic ambiguity signal. On LIBERO-Plus, ProGAL-VLA increases robustness under robot perturbations from 30.3 to 71.5 percent, reduces language ignorance by 3x-4x, and improves entity retrieval from 0.41 to 0.71 Recall@1. On the Custom Ambiguity Benchmark, it reaches AUROC 0.81 (vs. 0.52), AUPR 0.79, and raises clarification on ambiguous inputs from 0.09 to 0.81 without harming unambiguous success. The verification bottleneck increases mutual information of language-actions, the GAC loss imposes an entity-level InfoNCE bound, and attention entropy yields calibrated selective prediction, indicating that explicit verified grounding is an effective path toward instruction-sensitive, ambiguity-aware agents.
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。
