Co-Evolving Harnesses and Models: On-Policy Correction Helps Weaker Models Catch Up Where Imitation Fails Paper • 2609.09134 • Published 8 days ago • 10
DriveZero: End-to-End Driving Beyond Human Demonstrations Paper • 2609.06055 • Published 11 days ago • 63
Harnessing CLIP and DINO: An Uncertainty-Aware Cascaded Fusion Network for Generalizable Deepfake Image Detection Paper • 2609.07670 • Published 9 days ago • 18
Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal Paper • 2609.04482 • Published 13 days ago • 12
EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction Paper • 2609.02783 • Published 14 days ago • 120
SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers Paper • 2609.01343 • Published 15 days ago • 104