ProbeAct: Probe-Guided Training-Free Failure Recovery in Vision-Language-Action Models

• URL: http://arxiv.org/abs/2606.09740v1
• Date: Mon, 08 Jun 2026 17:04:24 GMT
• Title: ProbeAct: Probe-Guided Training-Free Failure Recovery in Vision-Language-Action Models
• Abstract summary: PROBEACT is a training-free runtime intervention frame-6 work.<n>It detects and recovers from grasping and placement failures in pre-7 trained VLA policies.
• Score: 23.3892541362594
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language-Action (VLA) models demonstrate strong perfor-1 mance on language-conditioned robotic manipulation within their training dis-2 tribution, yet their generalization capabilities remain fundamentally limited. They3 lack the robustness required to handle perturbations, frequently failing when con-4 fronted with lighting changes, altered camera viewpoints, or small initial-state5 variations. We propose PROBEACT, a training-free runtime intervention frame-6 work that detects and recovers from grasping and placement failures in pre-7 trained VLA policies without modifying their weights or requiring additional8 demonstrations. PROBEACT combines three components: (i) a lightweight multi-9 target hidden-state probe that predicts the 3D positions of task-relevant objects10 from intermediate VLA features, with Hungarian-matched identity tracking for11 multi-object scenes; (ii) an object-agnostic kinematic state machine that detects12 grasp, transport, and placement failures using only gripper-internal signals and13 end-effector kinematics; and (iii) a hierarchical Control Barrier Function (CBF)14 filter that encodes repeated-failure locations as soft safe-set constraints, mini-15 mally correcting VLA actions while preserving baseline behavior. As a plug-and-16 play, training-free intervention loop, PROBEACT is orthogonal to existing train-17 ing pipelines. Evaluated on the LIBERO-plus benchmark, our framework acts as18 a universal safety net, improving the success rate of the OpenVLA-OFT model19 from 69.6% to 74.1%, while demonstrating broad applicability to both base and20 fine-tuned VLA policies.

Related papers

• GloVLA: Let Geometry Move and Local VLA Interact for Robust Object-Centric Manipulation in Unstructured Environments [15.479388607422736]
  Vision-language-action (VLA) models have shown promising generalization for language-conditioned robot manipulation.<n>A single end-to-end VLA policy must simultaneously solve long-range transport of the end effector to task-relevant regions.<n>We introduce GloVLA, a hybrid framework that explicitly separates object-centric manipulation into two complementary regimes.
  arXiv  Detail & Related papers  (2026-09-05T20:48:14Z)
• GIFT: Guided Intermediate Feature Training via Action-Oriented Structural Supervision for Robotic Manipulation [60.95802802905149]
  We present GIFT (Guided Intermediate Feature Training), an architecture-flexible framework for learning intermediate features.<n>We instantiate GIFT in a Vision-Language-Action (VLA) policy, a direct-action World-Action Model (WAM), and an inverse-dynamics WAM while retaining each model's action formulation.<n>On RoboCasa, the three GIFT variants reach 61.4%, 83.6%, and 82.3%, outperforming their counterparts by 12.6, 9.0, and 8.4 points, respectively.
  arXiv  Detail & Related papers  (2026-09-03T17:59:03Z)
• PhysVLA: Towards Physically-Grounded VLA for Embodied Robotic Manipulation [2.7921261962314237]
  Vision-Language-Action (VLA) models excel at mapping visual inputs and natural language instructions directly to robotic control policies.<n>Because they are trained primarily to fit behavioural demonstration data, they do not explicitly enforce fundamental physical principles such as rigid-body dynamics or contact constraints.<n>This exposes a critical physics gap: standard temporal smoothing applied on top of single-step or chunked VLAs trades trajectory quality for added failures that short-term memory cannot resolve.<n>We introduce PhysVLA, a plug-and-play, inference-time framework designed to wrap any frozen VLA backbone without retraining, fine-
  arXiv  Detail & Related papers  (2026-06-11T20:23:09Z)
• Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models [25.34659221262947]
  Vision-Language-Action (VLA) models have demonstrated impressive end-to-end performance across a variety of robotic manipulation tasks.<n>Existing safety filters sidestep this problem by querying a vision-language model (VLM) to identify obstacles and their locations.<n>We discover that a small number of attention heads within a VLA model reliably localize the object the policy intends to approach.
  arXiv  Detail & Related papers  (2026-06-08T17:11:16Z)
• GEAR-VLA: Learning Geometry-Aware Action Representations for Generalizable Robotic Manipulation [49.16739604572808]
  Vision-Language-Action (VLA) models achieve strong benchmark performance but struggle in real-world deployment with unseen objects.<n>We argue that this stems from the lack of a unified geometry-aware manipulation representation.<n>We propose GEAR-VLA, a VLA framework for learning unified geometry-aware action representations for generalizable robotic manipulation.
  arXiv  Detail & Related papers  (2026-06-07T09:23:16Z)
• Mitigating Hallucination in Vision-Language Models through Barrier-Regulated Adaptive Closed-form Steering [2.497926557563177]
  Large vision-language models (LVLMs) often hallucinate objects that are not present in the input image.<n>Existing inference-time mitigation methods modify logits or hidden states throughout generation.<n>They lack an explicit grounding objective, intervene even when the model is already well-grounded, and use fixed correction strengths that do not adapt to the severity of grounding failure.<n>We propose BRACS, a training-free steering framework that addresses these issues through barrier-regulated adaptive closed-form steering.
  arXiv  Detail & Related papers  (2026-05-28T13:07:01Z)
• Self-Correcting VLA: Online Action Refinement via Sparse World Imagination [55.982504915794514]
  We propose Self-Correcting VLA (SC-VLA), which achieve self-improvement by intrinsically guiding action refinement through sparse imagination.<n>SC-VLA achieve state-of-the-art performance, yielding the highest task throughput with 16% fewer steps and a 9% higher success rate than the best-performing baselines.
  arXiv  Detail & Related papers  (2026-02-25T06:58:06Z)
• Universal Pose Pretraining for Generalizable Vision-Language-Action Policies [83.39008378156647]
  Existing Vision-Language-Action (VLA) models often suffer from feature collapse and low training efficiency.<n>We propose Pose-VLA, a decoupled paradigm that separates VLA training into a pre-training phase for extracting universal 3D spatial priors.<n>Our framework follows a two-stage pre-training pipeline, establishing fundamental spatial grounding via poses followed by motion alignment.
  arXiv  Detail & Related papers  (2026-02-23T11:00:08Z)
• On Robustness of Vision-Language-Action Model against Multi-Modal Perturbations [52.1029745126386]
  In vision-language-action (VLA) models, robustness to real-world perturbations is critical for deployment.<n>We propose RobustVLA against perturbations in VLA inputs and outputs.<n> Experiments on LIBERO demonstrate our RobustVLA delivers absolute gains over baselines of 12.6% on the pi0 backbone and 10.4% on the OpenVLA backbone.
  arXiv  Detail & Related papers  (2025-09-26T14:42:23Z)
• GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs [56.93583799109029]
  GrAInS is an inference-time steering approach that operates across both language-only and vision-language models and tasks.<n>During inference, GrAInS hidden activations at transformer layers guided by token-level attribution signals, and normalizes activations to preserve representational scale.<n>It consistently outperforms both fine-tuning and existing steering baselines.
  arXiv  Detail & Related papers  (2025-07-24T02:34:13Z)
• ConRFT: A Reinforced Fine-tuning Method for VLA Models via Consistency Policy [10.596344084789434]
  Vision-Language-Action (VLA) models have shown substantial potential in real-world robotic manipulation.<n>Fine-tuning these models through supervised learning struggles to achieve robust performance due to limited, inconsistent demonstrations.<n>We propose a reinforced fine-tuning approach for VLA models, named ConRFT, which consists of offline and online fine-tuning.
  arXiv  Detail & Related papers  (2025-02-08T05:01:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.