ExToken: Structured Exploration for Efficient Vision-Language-Action Reinforcement Fine-tuning

• URL: http://arxiv.org/abs/2607.12931v1
• Date: Tue, 14 Jul 2026 16:04:41 GMT
• Title: ExToken: Structured Exploration for Efficient Vision-Language-Action Reinforcement Fine-tuning
• Abstract summary: Reinforcement Learning (RL) has demonstrated significant potential for improving Vision-Language-Action (VLA) models on complex manipulation tasks.<n>In this work, we first investigate the exploration stagnation bottleneck in current VLA-RL frameworks.<n>We introduce RL Exploration Token (ExToken), a simple yet general framework that condition VLA policies on discrete behavioral priors.
• Score: 53.17987412112712
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement Learning (RL) has demonstrated significant potential for improving Vision-Language-Action (VLA) models on complex manipulation tasks. However, its practical scalability remains severely limited by the substantial cost of environmental interactions. In this work, we first investigate the exploration stagnation bottleneck in current VLA-RL frameworks and reveal that trajectory diversity is fundamentally more important to sample efficiency than the sheer quantity of collected rollouts. Motivated by these insights, we introduce RL Exploration Token (ExToken), a simple yet general framework that condition VLA policies on discrete behavioral priors derived from offline demonstrations for structured exploration. By conditioning the policy on different tokens during rollout collection, ExToken encourages the agent to explore diverse behavioral modes, substantially improving state-action coverage and exploration efficiency. To bridge exploration during training with deterministic inference at deployment, ExToken further incorporates a state-conditioned token selector that adaptively predicts effective behavioral modes for unseen scenarios. Extensive experiments across simulated and real-world robotic manipulation tasks demonstrate that ExToken consistently accelerates convergence, improves task performance, and exhibits strong robustness under highly constrained interaction budgets.

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)
• Learning to Watch: Active Video Anomaly Understanding via Interleaved Policy Optimization [56.150184676745205]
  Video anomaly understanding (VAU) relies on sparse, context-dependent cues.<n>We propose $Anomtext-$, a closed-loop framework that reconceptualizes video understanding as an active sequential decision-making process.<n>Inspired by human video-reviewing behavior, this framework unifies internal cognitive reasoning and strategic evidence acquisition into an interleaved policy.
  arXiv  Detail & Related papers  (2026-07-01T08:41:49Z)
• Nudging Beyond the Comfort Zone: Efficient Strategy-Guided Exploration for RLVR [53.27792011950384]
  Reinforcement learning with verifiable rewards (RLVR) has emerged as a scalable paradigm for improving the reasoning capabilities of large language models.<n>We propose NudgeRL, a framework for structured and diversity-driven exploration in RLVR.<n>Our approach introduces Strategy Nudging, which conditions each rollout on lightweight, strategy-level contexts.
  arXiv  Detail & Related papers  (2026-05-15T08:22:59Z)
• AIPO: Learning to Reason from Active Interaction [54.10819421625103]
  AIPO enables the policy model to proactively consult three functional collaborative agents when encountering reasoning bottlenecks.<n>AIPO consistently improves reasoning performance, generalizes robustly across different policy models and RLVR algorithms, and effectively expands the reasoning capability boundary of the policy model.
  arXiv  Detail & Related papers  (2026-05-08T19:06:55Z)
• Act, Think or Abstain: Complexity-Aware Adaptive Inference for Vision-Language-Action Models [7.802379200026965]
  We propose an adaptive framework that dynamically routes VLA execution based on the complexity of the perceived state.<n>Our approach transforms the VLA's vision-language backbone into an active detection tool by projecting latent embeddings into an ensemble of parametric and non-parametric estimators.
  arXiv  Detail & Related papers  (2026-03-05T13:14:41Z)
• SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models [21.133970394496327]
  Vision-Language-Action (VLA) models have emerged as a promising paradigm for general-purpose robotic control.<n>Current test-time scaling (TTS) methods require additional training, verifiers, and multiple forward passes, making them impractical for deployment.<n>We propose a simple inference strategy that jointly modulates visual perception and action based on'self-uncertainty'
  arXiv  Detail & Related papers  (2026-02-04T04:48:16Z)
• \ extsc{NaVIDA}: Vision-Language Navigation with Inverse Dynamics Augmentation [50.027425808733994]
  textscNaVIDA is a unified VLN framework that couples policy learning with action-grounded visual dynamics and adaptive execution.<n>textscNaVIDA augments training with chunk-based inverse-dynamics supervision to learn causal relationship between visual changes and corresponding actions.<n>Experiments show that textscNaVIDA achieves superior navigation performance compared to state-of-the-art methods with fewer parameters.
  arXiv  Detail & Related papers  (2026-01-26T06:16:17Z)
• DEAS: DEtached value learning with Action Sequence for Scalable Offline RL [46.40818333031899]
  Action Sequence (DEAS) is a simple yet effective offline RL framework that leverages action sequences for value learning.<n>DEAS consistently outperforms baselines on complex, long-horizon tasks from OGBench.<n>It can be applied to enhance the performance of large-scale Vision-Language-Action models.
  arXiv  Detail & Related papers  (2025-10-09T03:11:09Z)
• Agentic Jigsaw Interaction Learning for Enhancing Visual Perception and Reasoning in Vision-Language Models [63.69856480318313]
  AGILE formulates jigsaw solving as an interactive process, enabling the model to progressively engage with the environment.<n>We show that AGILE substantially boosts performance on jigsaw tasks of varying complexity.<n>We also demonstrate strong generalization across 9 general vision tasks, achieving an average improvement of 3.1%.
  arXiv  Detail & Related papers  (2025-10-01T17:58:05Z)

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.