Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures

• URL: http://arxiv.org/abs/2603.01063v1
• Date: Sun, 01 Mar 2026 11:41:22 GMT
• Title: Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures
• Abstract summary: 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.
• Score: 14.313346858887286
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language-Action (VLA) models for autonomous driving often hit a performance plateau during Reinforcement Learning (RL) optimization. This stagnation arises from exploration capabilities constrained by previous Supervised Fine-Tuning (SFT), leading to persistent failures in long-tail scenarios. In these critical situations, all explored actions yield a zero-value driving score. This information-sparse reward signals a failure, yet fails to identify its root cause -- whether it is due to incorrect planning, flawed reasoning, or poor trajectory execution. To address this limitation, we propose VLA with Explicit Learning from Failures (ELF-VLA), a framework that augments RL with structured diagnostic feedback. Instead of relying on a vague scalar reward, our method produces detailed, interpretable reports that identify the specific failure mode. The VLA policy then leverages this explicit feedback to generate a Feedback-Guided Refinement. By injecting these corrected, high-reward samples back into the RL training batch, our approach provides a targeted gradient, which enables the policy to solve critical scenarios that unguided exploration cannot. Extensive experiments demonstrate that our method unlocks the latent capabilities of VLA models, achieving state-of-the-art (SOTA) performance on the public NAVSIM benchmark for overall PDMS, EPDMS score and high-level planning accuracy.

Related papers

• RL$^2$-VLA: Adaptive RL Latent Compositional Steering with Test-Time Scaling for Vision-Language-Action Models [8.012549967082832]
  $RL2$ is an adaptive inference-time steering framework for Vision-Language-Action (VLA) models.<n>We train a lightweight offline RL policy conditioned on expressive latents extracted from the VLA action expert.<n>This compositional steering strategy combines the behavioral priors of large-scale imitation learning with the action diversity induced by offline RL.
  arXiv  Detail & Related papers  (2026-07-29T14:49:56Z)
• dVLA-RL: Reinforcement Learning over Denoising Trajectories for Discrete Diffusion Vision-Language-Action Models [49.497309561043004]
  We propose textbfdVLA-RL, shifting the learning objective from the marginal action probability to the joint probability of the sampled generation path.<n>Our approach achieves a success rate of textbf99.7% on LIBERO.<n>It also establishes strong VLA-based results on RoboTwin 2.0 by delivering a textbf30.6% improvement over the SFT baseline.
  arXiv  Detail & Related papers  (2026-06-22T17:19:03Z)
• Failing Forward: Adaptive Failure-Informed Learning for Vision-Language-Action Models [32.87104367896446]
  We propose an end-to-end framework that leverages failure trajectories as adaptive negative guidance for diffusion- and flow-based VLA policies.<n>AFIL uses a pretrained VLA to generate failure rollouts online, avoiding the need for handcrafted failure-mode design.<n>It then jointly trains Dual Action Generators (DAGs) for successful and failed behaviors while sharing a common vision-language backbone.
  arXiv  Detail & Related papers  (2026-05-08T19:57:11Z)
• Jump-Start Reinforcement Learning with Vision-Language-Action Regularization [1.2599533416395767]
  Reinforcement learning (RL) enables high-frequency, closed-loop control for robotic manipulation.<n>Current limitations hinder their direct use in fast and precise manipulation.<n>We propose Vision-Language-Action Jump-Starting (VLAJS) to improve exploration and learning efficiency.
  arXiv  Detail & Related papers  (2026-04-15T11:17:54Z)
• KEPO: Knowledge-Enhanced Preference Optimization for Reinforcement Learning with Reasoning [24.072603982041798]
  Reinforcement learning has emerged as a promising paradigm for inducing explicit reasoning behaviors in large language and vision-language models.<n>However, reasoning-oriented RL post-training remains fundamentally challenging due to sparse trajectory-level rewards.<n>Recent on-policy distillation methods introduce dense teacher supervision to stabilize optimization, but apply it uniformly across all generated trajectories.
  arXiv  Detail & Related papers  (2026-01-30T23:28:37Z)
• Evidence-Augmented Policy Optimization with Reward Co-Evolution for Long-Context Reasoning [52.144281362465996]
  We propose EAPO (Evidence-Augmented Policy Optimization) to apply Reinforcement Learning to long-context scenarios.<n>We first establish the Evidence-Augmented Reasoning paradigm, validating via Tree-Structured Evidence Sampling.<n>We then introduce a specialized RL algorithm where a reward model computes a Group-Relative Evidence Reward.<n>To sustain accurate supervision throughout training, we further incorporate an Adaptive Reward-Policy Co-Evolution mechanism.
  arXiv  Detail & Related papers  (2026-01-15T11:40:57Z)
• 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)
• Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning [71.19675094463834]
  This work introduces Counterfactual VLA, a self-reflective VLA framework that enables the model to reason about and revise its planned actions before execution.<n>CF-VLA first generates time-segmented meta-actions that summarize driving intent, and then performs counterfactual reasoning conditioned on both the meta-actions and the visual context.<n> Experiments on large-scale driving datasets show that CF-VLA improves trajectory accuracy by up to 17.6%, enhances safety metrics by 20.5%, and exhibits adaptive thinking.
  arXiv  Detail & Related papers  (2025-12-30T19:04:17Z)
• dVLM-AD: Enhance Diffusion Vision-Language-Model for Driving via Controllable Reasoning [69.36145467833498]
  We introduce dVLM-AD, a diffusion-based vision-language model that unifies perception, structured reasoning, and low-level planning for end-to-end driving.<n> evaluated on nuScenes and WOD-E2E, dVLM-AD yields more consistent reasoning-action pairs and achieves planning performance comparable to existing driving VLM/VLA systems.
  arXiv  Detail & Related papers  (2025-12-04T05:05:41Z)
• 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)
• Learning Affordances at Inference-Time for Vision-Language-Action Models [50.93181349331096]
  In robotics, Vision-Language-Action models (VLAs) offer a promising path towards solving complex control tasks.<n>We introduce Learning from Inference-Time Execution (LITEN), which connects a VLA low-level policy to a high-level VLM that conditions on past experiences.<n>Our approach iterates between a reasoning phase that generates and executes plans for the low-level VLA, and an assessment phase that reflects on the resulting execution.
  arXiv  Detail & Related papers  (2025-10-22T16:43:29Z)
• VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning [14.099306230721245]
  We present VLA-RL, an exploration-based framework that improves on online collected data at test time.<n>We fine-tune a pretrained vision-language model as a robotic process reward model, which is trained on pseudo reward labels annotated on automatically extracted task segments.<n>VLA-RL enables OpenVLA-7B to surpass the strongest finetuned baseline by 4.5% on 40 challenging robotic manipulation tasks in LIBERO.
  arXiv  Detail & Related papers  (2025-05-24T14:42:51Z)
• GTR: Guided Thought Reinforcement Prevents Thought Collapse in RL-based VLM Agent Training [62.536191233049614]
  Reinforcement learning with verifiable outcome rewards (RLVR) has effectively scaled up chain-of-thought (CoT) reasoning in large language models (LLMs)<n>This work investigates this problem through extensive experiments on complex card games, such as 24 points, and embodied tasks from ALFWorld.<n>We find that when rewards are based solely on action outcomes, RL fails to incentivize CoT reasoning in VLMs, instead leading to a phenomenon we termed thought collapse.
  arXiv  Detail & Related papers  (2025-03-11T15:17:02Z)
• Validity Learning on Failures: Mitigating the Distribution Shift in Autonomous Vehicle Planning [2.3558144417896583]
  The planning problem constitutes a fundamental aspect of the autonomous driving framework. We propose Validity Learning on Failures, VL(on failure) as a remedy to address this issue. We show that VL(on failure) outperforms the state-of-the-art methods by a large margin.
  arXiv  Detail & Related papers  (2024-06-03T17:25:18Z)

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.