Learning, Fast and Slow: Towards LLMs That Adapt Continually Paper • 2605.12484 • Published 12 days ago • 17
COMPASS: COntinual Multilingual PEFT with Adaptive Semantic Sampling Paper • 2604.20720 • Published Apr 22 • 2
KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning Paper • 2602.14293 • Published Feb 15 • 1
HybridStitch: Pixel and Timestep Level Model Stitching for Diffusion Acceleration Paper • 2603.07815 • Published Mar 8 • 10
Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization Paper • 2602.02958 • Published Feb 3 • 34
TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times Paper • 2512.16093 • Published Dec 18, 2025 • 97
Arbitrage: Efficient Reasoning via Advantage-Aware Speculation Paper • 2512.05033 • Published Dec 4, 2025 • 17
Understanding and Imitating Human-Robot Motion with Restricted Visual Fields Paper • 2410.05547 • Published Oct 7, 2024
PromptShield: Deployable Detection for Prompt Injection Attacks Paper • 2501.15145 • Published Jan 25, 2025