REAP the Experts: Why Pruning Prevails for One-Shot MoE compression Paper • 2510.13999 • Published Oct 15, 2025 • 10
Scaling Open-Ended Reasoning to Predict the Future Paper • 2512.25070 • Published about 1 month ago • 16
TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times Paper • 2512.16093 • Published Dec 18, 2025 • 95
Openhands Trajectories Collection Dataset of 67,074 OpenHands trajectories collected with Qwen3-Coder-480B-A35B-Instruct and two RFT checkpoints trained on the data • 3 items • Updated Dec 23, 2025 • 6
Next-Embedding Prediction Makes Strong Vision Learners Paper • 2512.16922 • Published Dec 18, 2025 • 85
VersatileFFN: Achieving Parameter Efficiency in LLMs via Adaptive Wide-and-Deep Reuse Paper • 2512.14531 • Published Dec 16, 2025 • 14
Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics Paper • 2512.12602 • Published Dec 14, 2025 • 44
QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management Paper • 2512.12967 • Published Dec 15, 2025 • 108
Bolmo Collection Artifacts for the Bolmo release: https://allenai.org/papers/bolmo. • 4 items • Updated Dec 23, 2025 • 12
Olmo 3.1 Collection The latest members of the Olmo 3 family: another 3 weeks of RL for 32B Think, the 32B Instruct model, large post-training research datasets... • 9 items • Updated Dec 23, 2025 • 47