LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling Paper • 2606.18023 • Published 5 days ago • 191
HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretraining Paper • 2606.20521 • Published 3 days ago • 5
Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance Paper • 2606.19195 • Published 4 days ago • 105
Learning from the Self-future: On-policy Self-distillation for dLLMs Paper • 2606.18195 • Published 5 days ago • 71
On the Scaling of PEFT: Towards Million Personal Models of Trillion Parameters Paper • 2606.02437 • Published 20 days ago • 231
Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs Paper • 2605.30611 • Published 24 days ago • 206
Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding Paper • 2605.29707 • Published 24 days ago • 147
CoHyDE: Iterative Co-Training of LLM Rewriter & Dense Encoder for Tool Retrieval Paper • 2605.29271 • Published 24 days ago • 9
EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation Paper • 2605.23271 • Published 30 days ago • 80