14 papers
Acoustic Tweezers for Magnetic Skyrmions
Chongzhou Wang, Weichao Yu
Current methods for driving magnetic skyrmions predominantly translate ensembles as a whole, lacking single-particle selectivity. Here, we propose an "acoustic tweezer" that determ…
Latent Reward Registers for Diffusion Preference Alignment
Yuanshen Guan, Zipeng Feng, Zhiwei Xiong +2
Aligning diffusion models with human preferences usually relies on a sparse terminal reward evaluated on the final generated samples, which creates a severe temporal credit-assignm…
X-Stage: An Overlooked Pipeline Stage for Communication-Computation Overlap in DiT Inference
Jianwen Xian, Zhiyuan Xu, Yuchen Li +8
Fine-grained, device-initiated communication lets persistent GPU kernels in distributed diffusion transformer (DiT) inference issue remote stores and overlap data movement with Ten…
HeadCast: Casting Attention Heads for Efficient Autoregressive Video Generation
Jinliang Shen, Lianghao Su, Zheming Li +4
Autoregressive (AR) video diffusion models have become a promising paradigm for long and streaming video synthesis, but the continuously growing Key-Value (KV) cache makes attentio…
JAGG: Jacobian-Aggregated Group Gradient for Efficient GRPO Training of Diffusion Models
Ruiyi Ding, Jie Li, He Kang +4
Group Relative Policy Optimization (GRPO) is a powerful reinforcement learning algorithm for aligning generative models with human preferences. While successful in large language m…
Kwai Summary Attention Technical Report
Chenglong Chu, Guorui Zhou, Guowang Zhang +35
Long-context ability, has become one of the most important iteration direction of next-generation Large Language Models, particularly in semantic understanding/reasoning, code agen…