Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA Models

• URL: http://arxiv.org/abs/2607.03751v1
• Date: Sat, 04 Jul 2026 07:53:10 GMT
• Title: Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA Models
• Abstract summary: Vision-Language-Action (VLA) models acquire broad embodied capabilities through large-scale pretraining.<n>The prevailing remedy, post-training, improves task-specific performance but narrows the generalist capability that makes pretraining valuable.<n>We propose SVA, a framework that equips frozen VLA policies with long-term consequence awareness.
• Score: 60.17417174639865
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language-Action (VLA) models acquire broad embodied capabilities through large-scale pretraining, yet their generalization remains far more fragile than that of LLMs and VLMs. The prevailing remedy, post-training via supervised fine-tuning or reinforcement learning, improves task-specific performance but narrows the generalist capability that makes pretraining valuable. We identify a key bottleneck: VLA failures stem not only from action generation but also from action evaluation. A diagnostic pass@k study confirms that frozen VLAs already contain competent behaviors in their output distribution, with overall success rates rising from 33% at pass@1 to 92% at pass@32. Inspired by this, we propose SVA (Search, Value, and Act), a simple framework that equips frozen VLA policies with long-term consequence awareness. SVA first uses Monte-Carlo tree search in simulation to fully explore the VLA's output distribution and collect diverse trajectories annotated with empirical returns; this knowledge is then distilled into a lightweight Q-value model that predicts the expected consequence of candidate actions; at deployment, the frozen VLA proposes multiple candidates and the evaluator selects the one with the highest uncertainty-regularized Q-value, requiring no simulator access. By decoupling action proposal from consequence evaluation, SVA preserves the generalization capacity of the VLA backbone while substantially improving task success rates. Experiments across embodied benchmarks show that SVA consistently improves generalization on unseen tasks and exhibits strong test-time scaling behavior. Strikingly, SVA enables a 9B VLA to outperform a 27B VLA by 7 points at 27% lower inference latency, suggesting that scaling test-time evaluation is more cost-effective than scaling model size.

Related papers

• VLAConf: Calibrated Task-Success Confidence for Vision-Language-Action Models [7.329348634749636]
  We propose VLAConf, a one-class discriminative confidence framework for Vision-Language-Action (VLA) models.<n>By leveraging frozen pretrained VLA internal representations, VLAConf directly estimates step-wise anomaly scores in a single forward pass.<n>Experiments on the LIBERO benchmark demonstrate that VLAConf significantly improves the quality of the confidence signal constructed for post-hoc calibration.
  arXiv  Detail & Related papers  (2026-05-28T08:42:12Z)
• QuoVLA: Quotient Space for Vision-Language-Action Models [51.02329790939691]
  Vision-Language-Action (VLA) models commonly adapt pretrained VLM latents to robot control by mapping visual observations and language instructions to continuous actions.<n>Our textitQuotient Theory for VLA shows that pretrained VLM latents are not action-insufficient but action-sufficient.<n>We propose QuoVLA, a quotient-space framework for VLA that compresses pretrained VLM latents into action-sufficient representations.
  arXiv  Detail & Related papers  (2026-05-24T06:28:53Z)
• Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures [14.313346858887286]
  Vision-Language-Action (VLA) models for autonomous driving often hit a performance plateau during Reinforcement Learning (RL) optimization.<n>This stagnation arises from exploration capabilities constrained by previous Supervised Fine-Tuning (SFT)<n>We propose Explicit Learning from Failures (ELF-VLA), a framework that augments RL with structured diagnostic feedback.
  arXiv  Detail & Related papers  (2026-03-01T11:41:22Z)
• When Vision Overrides Language: Evaluating and Mitigating Counterfactual Failures in VLAs [31.92520697946991]
  Vision-Language-Action models (VLAs) promise to ground language instructions in robot control, yet in practice often fail to faithfully follow language.<n>We show that counterfactual failures are prevalent yet underexplored across state-of-the-art VLAs.<n>We propose Counterfactual Action Guidance (CAG), a simple yet effective dual-branch inference scheme.
  arXiv  Detail & Related papers  (2026-02-19T18:59:20Z)
• CycleVLA: Proactive Self-Correcting Vision-Language-Action Models via Subtask Backtracking and Minimum Bayes Risk Decoding [60.06899554269808]
  CycleVLA is a system that equips Vision-Language-Action models (VLAs) with proactive self-correction.<n>CycleVLA achieves this by integrating a progress-aware VLA that flags critical subtask transition points.<n>Extensive experiments show that CycleVLA improves performance for both well-trained and under-trained VLAs.
  arXiv  Detail & Related papers  (2026-01-05T17:31:01Z)
• Value Vision-Language-Action Planning & Search [1.631000263754549]
  Vision-Language-Action (VLA) models have emerged as powerful generalist policies for robotic manipulation.<n>We introduce Value Vision-Language-Action Planning and Search (V-VLAPS), a framework that augments Monte Carlo Tree Search with a lightweight, learnable value function.<n>We evaluate V-VLAPS on the LIBERO robotic manipulation suite, demonstrating that our value-guided search improves success rates by over 5 percentage points.
  arXiv  Detail & Related papers  (2026-01-02T19:40:34Z)
• Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach [78.4812458793128]
  We propose textbfTACO, a test-time-scaling framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks.<n>Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits.
  arXiv  Detail & Related papers  (2025-12-02T14:42:54Z)
• DualVLA: Building a Generalizable Embodied Agent via Partial Decoupling of Reasoning and Action [62.70893433854428]
  We propose DualVLA, which enhances action performance through carefully designed post-training while still preserving reasoning capability.<n>Experiments show that DualVLA achieves an average success rate of 61.0 in SimplerEnv and an average score of 65.4 across eight competitive multimodal benchmarks.
  arXiv  Detail & Related papers  (2025-11-27T06:03:53Z)
• RoboMonkey: Scaling Test-Time Sampling and Verification for Vision-Language-Action Models [28.422082187079166]
  We introduce RoboMonkey, a test-time scaling framework for Vision-Language-Action (VLA) models.<n>RoboMonkey samples a small set of actions from a VLA, applies Gaussian perturbation and majority voting to construct an action proposal distribution, and then uses a Vision Language Model (VLM)-based verifier to select the optimal action.<n>We show that pairing existing VLAs with RoboMonkey yields significant performance gains, achieving a 25% absolute improvement on out-of-distribution tasks and 9% on in-distribution tasks.
  arXiv  Detail & Related papers  (2025-06-21T20:56:17Z)
• Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success [100.226572152954]
  We present an optimized fine-tuning recipe for vision-language-action models (VLAs)<n>Our recipe boosts OpenVLA's average success rate across four task suites from 76.5% to 97.1% while increasing action generation throughput by 26$times$.<n>In real-world evaluations, our fine-tuning recipe enables OpenVLA to successfully execute dexterous, high-frequency control tasks on a bimanual ALOHA robot.
  arXiv  Detail & Related papers  (2025-02-27T00:30:29Z)
• VL-RewardBench: A Challenging Benchmark for Vision-Language Generative Reward Models [66.56298924208319]
  Vision-language generative reward models (VL-GenRMs) play a crucial role in aligning and evaluating multimodal AI systems.<n>Current assessment methods primarily rely on AI-annotated preference labels from traditional tasks.<n>We introduce VL-RewardBench, a comprehensive benchmark spanning general multimodal queries, visual hallucination detection, and complex reasoning tasks.
  arXiv  Detail & Related papers  (2024-11-26T14:08:34Z)

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.