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
Robomonkey: Scaling test-time sampling and verifi- cation for vision-language-action models
Canonical reference. 100% of citing Pith papers cite this work as background.
22
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 5
citation-polarity summary
roles
background 5polarities
background 5representative citing papers
BOKBO is the first conformal abstention method for K-sample VLA policies that supplies finite-sample distribution-free guarantees on executed violation rates, with global and Mondrian per-task variants.
VeGAS improves MLLM-based embodied agents by sampling action ensembles and using a verifier trained on LLM-synthesized failure cases, yielding up to 36% relative gains on hard multi-object long-horizon tasks in Habitat and ALFRED.
UPS framework uses conformal prediction to calibrate VLM verifiers for choosing between high-confidence action execution, natural language task queries, or policy interventions, then applies residual learning from interventions to continually improve the base policy with minimal feedback.
ELASTIC learns a state-dependent meta-policy that allocates sequential and parallel test-time compute for generative control policies, outperforming fixed baselines in simulation and reducing real-robot latency by 34% at matched success.
SPARK reaches 43.7% success on six LIBERO-PRO cells by LLM-generated typed behavior trees plus multi-prompt perception and recovery, more than doubling CaP-Agent0 and VLA baselines.
E-TTS introduces a plug-and-play test-time scaling method for embodied tasks that unifies reasoning-action sampling with history buffers and closed-loop refinement to improve performance on manipulation benchmarks.
The log-probability ratio from RL post-training recovers the optimal advantage function, providing an effective free signal for test-time scaling, uncertainty estimation, and failure attribution in LLM agents.
DREAM-Chunk uses test-time sampling and latent-world-model rollouts to select robust action chunks from chunking-based VLA policies, improving performance under stochastic dynamics on simulation and hardware tasks.
VeriSpace is a 3D-aware action verifier that improves test-time action selection in VLA models by encoding scenes with visual and geometric information and reasoning over spatial relations and goal progress.
Q-VGM offline RL fine-tuning converts critic Q-gradients into residual velocity targets for flow-matching VLAs, raising LIBERO success from 75.0% to 92.5%.
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.
VLA-ATTC equips VLA models with adaptive test-time compute via an uncertainty clutch and relative action critic, cutting failure rates by over 50% on LIBERO-LONG.
FASTER models multi-candidate denoising as an MDP and trains a value function to filter actions early, delivering the performance of full sampling at lower cost in diffusion RL policies.
PDF improves VLA success rates on LIBERO and Atari by applying test-time perturbation learning with delayed feedback to correct trajectory overfitting and overconfidence.
A1 is a transparent VLA framework achieving state-of-the-art robot manipulation success with up to 72% lower latency via adaptive layer truncation and inter-layer flow matching.
Steerable VLAs trained on rich synthetic commands at subtask, motion, and pixel levels enable VLMs to steer robot behavior more effectively, outperforming prior hierarchical baselines on real-world manipulation and generalization tasks.
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
FORCE is a 3-stage RL fine-tuning method for VLA models that stabilizes Q-function via on-policy warm-up and filters high-value actions for updates, claiming 79% success rate gains and 32.5% faster training without human intervention.
TEXEDO uses test-time sampling and a combined feasibility-semantic reward model to select executable, text-aligned motions for humanoid robots from a pretrained generator.
A survey of test-time scaling for multimodal foundation models that introduces a three-way taxonomy of sampling, feedback, and search approaches along with applications and benchmarks.
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.
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.
-
BOKBO (Best of K Bad Options): Calibrated Abstention for VLA Policies
BOKBO is the first conformal abstention method for K-sample VLA policies that supplies finite-sample distribution-free guarantees on executed violation rates, with global and Mondrian per-task variants.
-
Think Twice, Act Once: Verifier-Guided Action Selection For Embodied Agents
VeGAS improves MLLM-based embodied agents by sampling action ensembles and using a verifier trained on LLM-synthesized failure cases, yielding up to 36% relative gains on hard multi-object long-horizon tasks in Habitat and ALFRED.
-
When to Act, Ask, or Learn: Uncertainty-Aware Policy Steering
UPS framework uses conformal prediction to calibrate VLM verifiers for choosing between high-confidence action execution, natural language task queries, or policy interventions, then applies residual learning from interventions to continually improve the base policy with minimal feedback.
-
ELASTIC: Efficiently Learning to Adaptively Scale Test-Time Compute for Generative Control Policies
ELASTIC learns a state-dependent meta-policy that allocates sequential and parallel test-time compute for generative control policies, outperforming fixed baselines in simulation and reducing real-robot latency by 34% at matched success.
-
Sequential Planning via Anchored Robotic Keypoints
SPARK reaches 43.7% success on six LIBERO-PRO cells by LLM-generated typed behavior trees plus multi-prompt perception and recovery, more than doubling CaP-Agent0 and VLA baselines.
-
E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulation
E-TTS introduces a plug-and-play test-time scaling method for embodied tasks that unifies reasoning-action sampling with history buffers and closed-loop refinement to improve performance on manipulation benchmarks.
-
Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents
The log-probability ratio from RL post-training recovers the optimal advantage function, providing an effective free signal for test-time scaling, uncertainty estimation, and failure attribution in LLM agents.
-
DREAM-Chunk: Reactive Action Chunking with Latent World Model
DREAM-Chunk uses test-time sampling and latent-world-model rollouts to select robust action chunks from chunking-based VLA policies, improving performance under stochastic dynamics on simulation and hardware tasks.
-
VeriSpace: Spatially Grounded Action Verification for Vision-Language-Action Models
VeriSpace is a 3D-aware action verifier that improves test-time action selection in VLA models by encoding scenes with visual and geometric information and reasoning over spatial relations and goal progress.
-
Q-VGM: Q-Value-Gradient Matching for Off-Policy Reinforcement Learning of Flow-Matching VLA
Q-VGM offline RL fine-tuning converts critic Q-gradients into residual velocity targets for flow-matching VLAs, raising LIBERO success from 75.0% to 92.5%.
-
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.
-
VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model
VLA-ATTC equips VLA models with adaptive test-time compute via an uncertainty clutch and relative action critic, cutting failure rates by over 50% on LIBERO-LONG.
-
FASTER: Value-Guided Sampling for Fast RL
FASTER models multi-candidate denoising as an MDP and trains a value function to filter actions early, delivering the performance of full sampling at lower cost in diffusion RL policies.
-
Test-Time Perturbation Learning with Delayed Feedback for Vision-Language-Action Models
PDF improves VLA success rates on LIBERO and Atari by applying test-time perturbation learning with delayed feedback to correct trajectory overfitting and overconfidence.
-
A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model
A1 is a transparent VLA framework achieving state-of-the-art robot manipulation success with up to 72% lower latency via adaptive layer truncation and inter-layer flow matching.
-
Steerable Vision-Language-Action Policies for Embodied Reasoning and Hierarchical Control
Steerable VLAs trained on rich synthetic commands at subtask, motion, and pixel levels enable VLMs to steer robot behavior more effectively, outperforming prior hierarchical baselines on real-world manipulation and generalization tasks.
-
A Survey on Vision-Language-Action Models for Embodied AI
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
-
FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation
FORCE is a 3-stage RL fine-tuning method for VLA models that stabilizes Q-function via on-policy warm-up and filters high-value actions for updates, claiming 79% success rate gains and 32.5% faster training without human intervention.
-
TEXEDO : Test Time Scaling for Controller-aware Language-conditioned Humanoid Motion Generation
TEXEDO uses test-time sampling and a combined feasibility-semantic reward model to select executable, text-aligned motions for humanoid robots from a pretrained generator.
-
Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning
A survey of test-time scaling for multimodal foundation models that introduces a three-way taxonomy of sampling, feedback, and search approaches along with applications and benchmarks.
-
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.