SARL optimizes language prompt inputs to generalist vision-language-action policies through online RL to solve complex long-horizon tasks by composing existing skills.
hub
Improving vision-language-action model with online reinforcement learning.arXiv preprint arXiv:2501.16664
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
hub tools
citation-role summary
background 3
other 1
citation-polarity summary
representative citing papers
PCM uses success-failure action variance to probabilistically select and mask chunks for gradient updates in GRPO, matching standard success rates with 2.38x wall-clock speedup and 60% lower memory on LIBERO benchmarks.
PhySe-RPO enables diffusion-based surgical smoke removal by converting restoration into a stochastic policy optimized with physics consistency and CLIP semantic rewards under limited supervision.
Success Visitation Matching uses a discriminator to turn sparse outcome rewards into dense process rewards by matching visitations of successful episodes, provably preserving the optimal policy and speeding up robotic RL finetuning.
Agentic-VLA enables efficient online adaptation of VLA models, delivering +12.3% on long-horizon tasks, +28.5% in 1-shot learning, and 2.4x faster convergence on LIBERO through three new components.
Veo-3 video predictions enable approximate task-level robot trajectories in zero-shot settings but require hierarchical integration with low-level VLA policies for reliable manipulation performance.
LifeLong-RFT applies chunking-level on-policy reinforcement learning with Quantized Action Consistency Reward, Continuous Trajectory Alignment Reward, and Format Compliance Reward to fine-tune VLA models, achieving a 22% average success rate gain over supervised fine-tuning on the LIBERO benchmark's
TwinRL expands RL exploration via digital twin reconstruction and twin RL warm-up to guide real-world learning, reaching near-100% success with 20 minutes of on-robot time across four tasks.
RECAP enables a generalist VLA to self-improve via advantage-conditioned RL on mixed real-world data, more than doubling throughput and halving failure rates on hard manipulation tasks.
A controllable world model trained on the DROID dataset generates consistent multi-view robot trajectories for over 20 seconds and improves generalist policy success rates by 44.7% via imagined trajectory fine-tuning.
SimpleVLA-RL applies tailored reinforcement learning to VLA models, reaching SoTA on LIBERO, outperforming π₀ on RoboTwin, and surpassing SFT in real-world tasks while reducing data needs and identifying a 'pushcut' phenomenon.
VLA-RL applies online RL to pretrained VLAs, yielding a 4.5% gain over strong baselines on 40 LIBERO manipulation tasks and matching commercial models like π₀-FAST.
DexVLA combines a scaled diffusion action expert with embodiment curriculum learning to achieve better generalization and performance than prior VLA models on diverse robot hardware and long-horizon tasks.
Video Prediction Policy conditions robot action learning on future-frame predictions inside fine-tuned video diffusion models, yielding 18.6% relative gains on Calvin ABC-D and 31.6% higher real-world success rates.
AllDayNav encodes scene dynamics into a large model's parameters via RL and a multimodal memory, achieving near-100% success rates in lifelong navigation and outperforming map-based and VLM baselines.
DyGRO-VLA is a two-stage optimization framework for cross-task scaling of Vision-Language-Action models via dynamic grouped residual optimization in RL.
Degradation-Driven Prompting improves VLM accuracy on deceptive visual tasks by downsampling inputs and routing them through classification, external visual tools, and a critic.
VGAS uses best-of-N selection with a geometrically grounded critic and explicit regularization to improve success rates of few-shot VLA policies under limited data and distribution shifts.
Reflective Self-Adaptation combines failure-reflective reinforcement learning with success-guided imitation learning to enable faster and more reliable task adaptation for pre-trained Vision-Language-Action models.
citing papers explorer
-
Adapting Generalist Robot Policies with Semantic Reinforcement Learning
SARL optimizes language prompt inputs to generalist vision-language-action policies through online RL to solve complex long-horizon tasks by composing existing skills.
-
Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking
PCM uses success-failure action variance to probabilistically select and mask chunks for gradient updates in GRPO, matching standard success rates with 2.38x wall-clock speedup and 60% lower memory on LIBERO benchmarks.
-
PhySe-RPO: Physics and Semantics Guided Relative Policy Optimization for Diffusion-Based Surgical Smoke Removal
PhySe-RPO enables diffusion-based surgical smoke removal by converting restoration into a stochastic policy optimized with physics consistency and CLIP semantic rewards under limited supervision.
-
Learning Process Rewards via Success Visitation Matching for Efficient RL
Success Visitation Matching uses a discriminator to turn sparse outcome rewards into dense process rewards by matching visitations of successful episodes, provably preserving the optimal policy and speeding up robotic RL finetuning.
-
Agentic-VLA: Efficient Online Adaptation for Vision-Language-Action Models
Agentic-VLA enables efficient online adaptation of VLA models, delivering +12.3% on long-horizon tasks, +28.5% in 1-shot learning, and 2.4x faster convergence on LIBERO through three new components.
-
Veo-Act: How Far Can Frontier Video Models Advance Generalizable Robot Manipulation?
Veo-3 video predictions enable approximate task-level robot trajectories in zero-shot settings but require hierarchical integration with low-level VLA policies for reliable manipulation performance.
-
Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning
LifeLong-RFT applies chunking-level on-policy reinforcement learning with Quantized Action Consistency Reward, Continuous Trajectory Alignment Reward, and Format Compliance Reward to fine-tune VLA models, achieving a 22% average success rate gain over supervised fine-tuning on the LIBERO benchmark's
-
TwinRL: Digital Twin-Driven Reinforcement Learning for Real-World Robotic Manipulation
TwinRL expands RL exploration via digital twin reconstruction and twin RL warm-up to guide real-world learning, reaching near-100% success with 20 minutes of on-robot time across four tasks.
-
$\pi^{*}_{0.6}$: a VLA That Learns From Experience
RECAP enables a generalist VLA to self-improve via advantage-conditioned RL on mixed real-world data, more than doubling throughput and halving failure rates on hard manipulation tasks.
-
Ctrl-World: A Controllable Generative World Model for Robot Manipulation
A controllable world model trained on the DROID dataset generates consistent multi-view robot trajectories for over 20 seconds and improves generalist policy success rates by 44.7% via imagined trajectory fine-tuning.
-
SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning
SimpleVLA-RL applies tailored reinforcement learning to VLA models, reaching SoTA on LIBERO, outperforming π₀ on RoboTwin, and surpassing SFT in real-world tasks while reducing data needs and identifying a 'pushcut' phenomenon.
-
VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning
VLA-RL applies online RL to pretrained VLAs, yielding a 4.5% gain over strong baselines on 40 LIBERO manipulation tasks and matching commercial models like π₀-FAST.
-
DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control
DexVLA combines a scaled diffusion action expert with embodiment curriculum learning to achieve better generalization and performance than prior VLA models on diverse robot hardware and long-horizon tasks.
-
Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations
Video Prediction Policy conditions robot action learning on future-frame predictions inside fine-tuned video diffusion models, yielding 18.6% relative gains on Calvin ABC-D and 31.6% higher real-world success rates.
-
AllDayNav: Lifelong Navigation via Real-World Reinforcement Learning
AllDayNav encodes scene dynamics into a large model's parameters via RL and a multimodal memory, achieving near-100% success rates in lifelong navigation and outperforming map-based and VLM baselines.
-
DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimization
DyGRO-VLA is a two-stage optimization framework for cross-task scaling of Vision-Language-Action models via dynamic grouped residual optimization in RL.
-
E-VLA: Event-Augmented Vision-Language-Action Model for Dark and Blurred Scenes
Degradation-Driven Prompting improves VLM accuracy on deceptive visual tasks by downsampling inputs and routing them through classification, external visual tools, and a critic.
-
VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action Adaptation
VGAS uses best-of-N selection with a geometrically grounded critic and explicit regularization to improve success rates of few-shot VLA policies under limited data and distribution shifts.
-
Reflection-Based Task Adaptation for Self-Improving VLA
Reflective Self-Adaptation combines failure-reflective reinforcement learning with success-guided imitation learning to enable faster and more reliable task adaptation for pre-trained Vision-Language-Action models.