Accepted Papers
List of accepted papers at the 8th Robot Learning Workshop @ NeurIPS 2026.
All accepted papers are available on OpenReview.
-
Augmentron: Scalable Multi-View Visual Augmentation for Robot Learning
-
ShadowVLA: Test-Time Re-Grounding of Frozen VLA Policies from Self-Generated References
-
Tiny Recursive Policy: Recursive Latent Refinement for Visuomotor Control
-
One Checkpoint, Three Robots, One Day: Zero-Shot Navigation as a System Property
-
SARAM: Sparse Adapter Allocation with Reuse and Merging for Lifelong Robot Learning
-
PhysCaP: Grounding Code-as-Policy Agent with Physics-Informed Exploration
-
Warp RL: Reshaping Base Policy Distributions for Dynamics Adaptation
-
Reliability-Aware Foundation-Model Supervision for Label-Efficient Robotic Perception
-
Chunking the Policy: Extending the Policy-Update Horizon with a Compact Transformer
-
GAINS: Leveraging Inconsistent Human Intervention Signals in Reinforcement Learning
-
Few-Shot Residual Adaptation of Pretrained Latent World Models
-
Enhancing VLM Reward Models Through Structure-Aware Fine-Tuning
-
Control-Aware Command Generation for Asynchronous Control and Action Inference with Vision-Based Policies towards Fast Execution
-
Democratizing Data-Efficient Robot Learning with Robot-Free Demonstrations
-
Count Generalization Is Easy, Actually: Warp a Frozen Skill Instead of Retraining It
-
Code Agents Automate the Two Manual Bottlenecks of Sim-Based Robot Learning: Digital Twin Scene Construction and Demonstration Collection
-
FactoryBench: Evaluating Industrial Machine Understanding
-
Video models are zero-shot learners and reasoners
-
Wild-World: Spatially Grounded Action-Conditioned RGB-D-Semantic World Modeling for Field Robotics
-
Silent Semantic Failures Below the Policy: A Historical Benchmark for Robot-Learning Software
-
How Do World Action Models Use Their Imagined Futures? Evidence from Mechanistic Interventions
-
DTG-VLA: Dynamic Target Guidance for Task-Compositional Generalization in Robot Manipulation
-
What Crosses Time? From Recurrent State to Causal Memory in Vision-Language-Action Models
-
Uncertainty-guided Next Best Iso-Surface Optimization for Neural Volumetric Capture
-
Shared Experiences for Collective World Models: Exact Team and Tree Synchronization Laws after Unpaired Learning
-
ENCORE: Few-Shot Agentic Discovery of Manipulation Abstractions
-
Task-Aware Conformal State Estimation for Reliable Robotic Planning
-
SCOUT: Per-Context Reset Curricula for Sparse-Reward Reinforcement Learning
-
Defining Operational Conditions for Safety-Critical AI-Based Systems from Data
-
SafeCine: Risk-Calibrated Physical Preference Supervision for Language-Conditioned Aerial Cinematography
-
SurVAL: Rollout-Free Survival Validation for Robot Policies
-
Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching
-
Did the World Change or Did My Sensor Break? Cheap Physics Residuals Attribute Deployment Faults for Frozen Manipulation Policies
-
Execution Aware VLAs for Robust Manipulation
-
ValueFormer: Stage-Aware Value Labels for Online Failure Detection and VLA Post-Training
-
CoHuB: A Collaborative Humanoid Loco-Manipulation Benchmark
-
Learning Actions, Not Noise: Direct Visuomotor Policy Learning on Action Manifolds
-
Adaptive Action Horizon Selection via Flow-Field Uncertainty
-
Adjoint Guidance Flow for Efficient Critic-Guided Vision-Language-Action Models
-
Subliminal-Style Action Traits in Vision-Language-Action Robot Policies
-
Learning Robot Behavior Priors from Non-Robotic Synthetic Data
-
PonderPounce: A Pretrained MLLM as an Episode Context Engine for Robot Control
-
Replay Buffer Distribution Dynamics Matter in Massively Parallel Off-Policy RL