SERA: Scale-Equalized Rollout Allocation for Maximum Likelihood Reinforcement LearningResearchSep 30, 2026
Compiled Agency: Coding Agents as Game AI Researchers -- from a Roguelike to StarCraft II and CivilizationResearchSep 30, 2026
Adapting Context Compression for Long-Horizon Agents with Counterfactual ContinuationsResearchSep 30, 2026
GA-EIRFS: A Geometry-Augmented Repeat-Factor Sampling Method for Long-Tailed LiDAR 3D Object DetectionResearchSep 30, 2026