Flow Reversal Steering steers flow matching generalist policies by reversing suboptimal actions to nearby better modes, enabling improved zero-shot control, quick distillation, and RL bootstrapping in robotic manipulation.
hub Canonical reference
A vision- language-action-critic model for robotic real-world rein- forcement learning.arXiv preprint arXiv:2509.15937
Canonical reference. 71% of citing Pith papers cite this work as background.
26
Pith papers citing it
Background
71% of classified citations
hub tools
citation-role summary
background 5
method 1
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.
ViVa turns a video generator into a value model for robot RL that jointly forecasts future states and task value, yielding better performance on real-world box assembly when integrated with RECAP.
STEAM learns advantages from expert trajectories via self-supervised temporal ensemble modeling to improve policy learning on real robot tasks like bimanual folding and pick-and-place.
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.
RARM is a lightweight visual comparator trained once on general videos that supplies dense progress rewards to RL by matching rollout clips to a reference demonstration and gating rewards on match confidence.
Fine-tuning VLMs with pairwise progress supervision from policy rollouts improves fine-grained failure detection and boosts robot manipulation success by 11% real-world and 5.9% in simulation.
Feat2Go uses patch-level similarity from a visual world model and trend-based clustering to create progress targets for training value models that improve reward shaping in embodied RL for VLA policies, yielding large gains on ManiSkill3 and RoboTwin benchmarks.
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.
A retrieve-then-steer method stores successful robot actions in memory and uses them to steer a frozen VLA's flow-matching sampler for better test-time reliability without parameter updates.
ARM trains reward models on Progressive/Regressive/Stagnant labels to enable adaptive reweighting in offline RL, reaching 99.4% success on towel-folding with minimal human intervention.
Robometer combines intra-trajectory progress supervision with inter-trajectory preference supervision on a 1M-trajectory dataset to learn more generalizable robotic reward functions than prior methods.
RISE combines a controllable dynamics model and progress value model into a closed-loop self-improving pipeline that updates robot policies entirely in imagination, reporting over 35% absolute gains on three real-world tasks.
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
PALM improves long-horizon robotic manipulation success by distilling affordance representations for object interaction and predicting within-subtask progress in a VLA model.
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.
InSight enables autonomous acquisition of manipulation primitives in VLAs via automated segmentation for steerability and a VLM-guided data flywheel that generates and integrates new demonstrations for tasks like pouring and sweeping.
World Value Model (WVM) integrates world models with value estimation to achieve SOTA Value-Order Correlation on expert and suboptimal robotic data and improves downstream policy performance.
HiL-ResRL trains a model-agnostic residual policy on VLA actions using human-guided online RL, achieving over 95% success rate after 1.5 hours of real-robot training.
DexPIE improves dexterous manipulation success rates by 37% over demo policies via real-world experience collection with adapted intervention, multi-stage DAgger, asynchronous relative-action inference, and optimality conditioning.
TapSampling improves generalist robotic manipulation policies at inference time via latent action sampling with an Action-VAE and selection by a task-progress outcome predictor.
Embodied reward models systematically over-reward unsafe, suboptimal, and shortcut robot behaviors due to training on successful data only, and modest inclusion of bad behavior data improves alignment with human preferences.
CaB predicts Before/Hit/After boundary tokens to produce auditable switching decisions and boundary-stable control in VLA agents on a Minecraft benchmark under single global calibration.
ABot-Claw is an embodied software layer that adds unified robot scheduling, cross-embodiment visual memory, and critic-driven replanning on top of OpenClaw to support persistent multi-robot execution from natural-language goals.
citing papers explorer
-
Improving Robotic Generalist Policies via Flow Reversal Steering
Flow Reversal Steering steers flow matching generalist policies by reversing suboptimal actions to nearby better modes, enabling improved zero-shot control, quick distillation, and RL bootstrapping in robotic manipulation.
-
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.
-
ViVa: A Video-Generative Value Model for Robot Reinforcement Learning
ViVa turns a video generator into a value model for robot RL that jointly forecasts future states and task value, yielding better performance on real-world box assembly when integrated with RECAP.
-
STEAM: Self-Supervised Temporal Ensemble Advantage Modeling for Real-World Robot Learning
STEAM learns advantages from expert trajectories via self-supervised temporal ensemble modeling to improve policy learning on real robot tasks like bimanual folding and pick-and-place.
-
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.
-
RARM: Confidence-Gated Progress Reward Modeling for RL in Manipulation
RARM is a lightweight visual comparator trained once on general videos that supplies dense progress rewards to RL by matching rollout clips to a reference demonstration and gating rewards on match confidence.
-
Robot Critics that Sweat the Small Stuff
Fine-tuning VLMs with pairwise progress supervision from policy rollouts improves fine-grained failure detection and boosts robot manipulation success by 11% real-world and 5.9% in simulation.
-
Feat2Go: Visual Feature-Grounded Value Estimation for Embodied Reinforcement Learning
Feat2Go uses patch-level similarity from a visual world model and trend-based clustering to create progress targets for training value models that improve reward shaping in embodied RL for VLA policies, yielding large gains on ManiSkill3 and RoboTwin benchmarks.
-
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.
-
Retrieve-then-Steer: Online Success Memory for Test-Time Adaptation of Generative VLAs
A retrieve-then-steer method stores successful robot actions in memory and uses them to steer a frozen VLA's flow-matching sampler for better test-time reliability without parameter updates.
-
ARM: Advantage Reward Modeling for Long-Horizon Manipulation
ARM trains reward models on Progressive/Regressive/Stagnant labels to enable adaptive reweighting in offline RL, reaching 99.4% success on towel-folding with minimal human intervention.
-
Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons
Robometer combines intra-trajectory progress supervision with inter-trajectory preference supervision on a 1M-trajectory dataset to learn more generalizable robotic reward functions than prior methods.
-
RISE: Self-Improving Robot Policy with Compositional World Model
RISE combines a controllable dynamics model and progress value model into a closed-loop self-improving pipeline that updates robot policies entirely in imagination, reporting over 35% absolute gains on three real-world tasks.
-
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
-
PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation
PALM improves long-horizon robotic manipulation success by distilling affordance representations for object interaction and predicting within-subtask progress in a VLA model.
-
$\pi^{*}_{0.6}$: 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.
-
InSight: Self-Guided Skill Acquisition via Steerable VLAs
InSight enables autonomous acquisition of manipulation primitives in VLAs via automated segmentation for steerability and a VLM-guided data flywheel that generates and integrates new demonstrations for tasks like pouring and sweeping.
-
World Value Models for Robotic Manipulation
World Value Model (WVM) integrates world models with value estimation to achieve SOTA Value-Order Correlation on expert and suboptimal robotic data and improves downstream policy performance.
-
HiL-ResRL: A Model-Agnostic Finetuning Adapter via Human-in-the-loop Residual Reinforcement Learning
HiL-ResRL trains a model-agnostic residual policy on VLA actions using human-guided online RL, achieving over 95% success rate after 1.5 hours of real-robot training.
-
DexPIE: Stable Dexterous Policy Improvement from Real-World Experience
DexPIE improves dexterous manipulation success rates by 37% over demo policies via real-world experience collection with adapted intervention, multi-stage DAgger, asynchronous relative-action inference, and optimality conditioning.
-
TapSampling: Inference-Time Sampling with a Task-Progress-Understanding Verifier for Robotic Manipulation
TapSampling improves generalist robotic manipulation policies at inference time via latent action sampling with an Action-VAE and selection by a task-progress outcome predictor.
-
Position: Good Embodied Reward Models Need Bad Behavior Data
Embodied reward models systematically over-reward unsafe, suboptimal, and shortcut robot behaviors due to training on successful data only, and modest inclusion of bad behavior data improves alignment with human preferences.
-
Completion at the Boundary (CaB): Deployable Switching with Completion-Aware Control under Limited Calibration
CaB predicts Before/Hit/After boundary tokens to produce auditable switching decisions and boundary-stable control in VLA agents on a Minecraft benchmark under single global calibration.
-
ABot-Claw: A Foundation for Persistent, Cooperative, and Self-Evolving Robotic Agents
ABot-Claw is an embodied software layer that adds unified robot scheduling, cross-embodiment visual memory, and critic-driven replanning on top of OpenClaw to support persistent multi-robot execution from natural-language goals.
- RoboProcessBench: Benchmarking Process-Aware Understanding in Vision-Language Robotic Manipulation
- GS-Playground: A High-Throughput Photorealistic Simulator for Vision-Informed Robot Learning