acceptodds: the papers of ICLR 2027
355 results for “area:robotics rating>=3” in ICLR 2027. 11 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]P-BFM: Perceptive Behavioral Foundation Models for Humanoid Control via Unsupervised Reinforcement Learning▲233.33
[PDF]SpeakingRobot: Semantic-Event-Driven Robot-Space Co-Speech Generation with Lower-Body Coordination▲1.7k3.33
[PDF]PhysDrive: A Physically Grounded Vision-Language Model for Driving Understanding and Planning▲5.3k3.00
[PDF]EgoMatrix: Physically Faithful Dexterous Robot Demonstrations from Egocentric Human Videos▲5.3k3.00
[PDF]VLA-Hallucination: Diagnosing and Mitigating Hallucination Failures in Vision-Language-Action Models▲5.3k3.00
[PDF]SocioWM: Social-Conditioned World Model for Social Context Vision-and-Language Navigation▲4643.00
[PDF]Learn the Operator, Not the Future: State-Anchored Distillation of World Action Model Dynamics for Autonomous Driving▲4.7k3.00
[PDF]Fine-Tuning VLAs with Self-Demonstrated Generative Control for Multi-Task Manipulation▼1.2k3.00
[PDF]The Action Doesn't End the Thought: Continuing Computation in Vision-Language-Action Models▼1.2k3.00
[PDF]Squeeze to Plan: Learning Dense Visual Representations for Efficient Planning via Bisimulation Information Bottleneck▲5.3k3.00
[PDF]Redundancy or Relevance: What Language-Guided Token Pruning Selects in Vision-Language-Action Models▲4643.00
[PDF]Bilinear World Models: Learning Representations with Structured Dynamics for Efficient Control.▲233.33
Withdrawn or desk rejected (11)
[PDF]Diagnosing the Structure and Local Behavioral Effects of Viewpoint-Induced Action Distortion in Robot PoliciesWithdrawn3.33
[PDF]Learning from Consequences: Posterior-Transition Reweighting for Behavior CloningDesk rejected3.00
[PDF]BeyondRetarget: Learning Executable Humanoid Motions Directly from Monocular VideoDesk rejected3.33
[PDF]Temporal Ensembling Revisited: Recovering the Stochastic Headroom of Generative Robot PoliciesDesk rejected3.00
[PDF]SmartVLA: A Steerable Action Interface for Fine-Grained, Multimodal, and Corrective InstructionsDesk rejected3.00