NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
843 results for “keyword:"world models"” in ICLR 2027. 295 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]WARM: A World Action Rectification Model Unifying VLA and WAM through a Planning-Rectification Architecture▼4.3k2.00
[PDF]Learning What Actions Change: Action Effect Matching for Dynamics-Faithful World Models▲3952.67
[PDF]Structured World Models for Sample-Efficient Planning in Partially Known Dynamical Systems▼4.5k2.67
[PDF]ActForesight: Predictive Safety for Embodied Agents with Action-Conditioned Visual Futures and Motion Monitoring▼1.2k1.67
[PDF]Learning State Transitions Improves Downstream Policy Learning in Geometry Theorem Proving▼13k2.00
[PDF]What-If World: A Contrastive Reasoning Benchmark for Video World Models in Embodied Scenarios▼4.5k2.67
[PDF]Same States, Different Actuators: What Single-Actuator Data Must Cover for Compositional World Models▲6.1k2.00
[PDF]DIM-WM: Factorized World Modeling for Platform Information Services in AI-Agent Data Markets▼5.8k2.33
[PDF]Hierarchical Reinforcement Learning with Temporally-extended Latent Action World Models▼4.3k2.00
[PDF]ANTE: Anticipate, Then Execute - Learning Internal World Models for Code Agents from Verified Beliefs▼5.8k2.33
[PDF]From Latent Loops to Better Navigation: Understanding Diffusion-Based World Models with Cognitive-Map-Inspired Dynamics▼5.8k2.33
[PDF]Second Thoughts: Exploring and Planning through Comparative Imagination in World Models▲3952.67
Withdrawn or desk rejected (295)
[PDF]What Should I Try Next? Decision-Relevant Counterfactual Information Gain for Active World ModelsWithdrawn1.00
[PDF]Defect Responses in Visual World Models: Long-Horizon Propagation, Gradient Interaction, and Planning OutcomesWithdrawn1.33
[PDF]Correlation Does Not Certify a Policy Choice: What Benchmarks Should Publish InsteadWithdrawn2.33
[PDF]Auto-JEPA: A Latent World Model of Continuous Intent for End-to-End Autonomous DrivingWithdrawn1.67