ReStruct steers robot policies at inference time by reconfiguring task structure with neural automata and synchronous products, claiming up to 25% gains over VLA models in success and preference adherence.
From foresight to forethought: Vlm-in-the-loop policy steering via latent alignment
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
years
2026 9verdicts
UNVERDICTED 9representative citing papers
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.
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.
APT pretrains the action expert as a vision-action prior on frozen VLM features then adds language through gated fusion to improve OOD instruction generalization in continuous-action VLA policies.
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.
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 training-free fusion layer enables stale VLM selections to improve a real-time planner's trajectory scoring for urban sidewalk navigation, yielding 30% ADE reduction in challenging scenarios.
VLESA introduces a goal-conditioned safety Q-filter trained via GRPO on egocentric video plus an intent-action predictor, achieving higher intervention accuracy and over 41 percentage points better action safety on the ASIMOV-2.0 benchmark.
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
-
Inference-Time Robot Behavior Steering through Physically-Aware Reconfiguration of Task-Structure
ReStruct steers robot policies at inference time by reconfiguring task structure with neural automata and synchronous products, claiming up to 25% gains over VLA models in success and preference adherence.
-
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.
-
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.
-
APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
APT pretrains the action expert as a vision-action prior on frozen VLM features then adds language through gated fusion to improve OOD instruction generalization in continuous-action VLA policies.
-
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.
-
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.
-
Slow Brain, Fast Planner: Latency-Resilient VLM-Augmented Urban Navigation
A training-free fusion layer enables stale VLM selections to improve a real-time planner's trajectory scoring for urban sidewalk navigation, yielding 30% ADE reduction in challenging scenarios.
-
VLESA: Vision-Language Embodied Safety Agent for Human Activity Monitoring
VLESA introduces a goal-conditioned safety Q-filter trained via GRPO on egocentric video plus an intent-action predictor, achieving higher intervention accuracy and over 41 percentage points better action safety on the ASIMOV-2.0 benchmark.
-
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.