TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models

TL;DR

This paper introduces TAG, a guidance mechanism for vision-language-action policies that enhances object grounding accuracy in cluttered scenes, significantly improving robotic manipulation robustness without altering existing models.

Contribution

We propose TAG, a simple inference-time guidance method inspired by classifier-free guidance, that reduces distractor bias in VLA policies without requiring architecture changes.

Findings

TAG improves robustness in cluttered scenes

Reduces near-miss and wrong-object actions

Enhances performance on manipulation benchmarks

Abstract

Vision--Language--Action (VLA) policies have shown strong progress in mapping language instructions and visual observations to robotic actions, yet their reliability degrades in cluttered scenes with distractors. By analyzing failure cases, we find that many errors do not arise from infeasible motions, but from instance-level grounding failures: the policy often produces a plausible grasp trajectory that lands slightly off-target or even on the wrong object instance. To address this issue, we propose TAG (Target-Agnostic Guidance), a simple inference-time guidance mechanism that explicitly reduces distractor- and appearance-induced bias in VLA policies. Inspired by classifier-free guidance (CFG), TAG contrasts policy predictions under the original observation and an object-erased observation, and uses their difference as a residual steering signal that strengthens the influence of object evidence in the decision process. TAG does not require modifying the policy architecture and can be integrated with existing VLA policies with minimal training and inference changes. We evaluate TAG on standard manipulation benchmarks, including LIBERO, LIBERO-Plus, and VLABench, where it consistently improves robustness under clutter and reduces near-miss and wrong-object executions.