activity
20242026
collaborators

7 papers

cs.CV2026

FVAttn: Adaptive Sparse Attention with Runtime Load Balancing for Video Generation

Hao Liu, Chenghuan Huang, Ye Huang +6

Video Diffusion Transformers process long spatio-temporal sequences, making self-attention the main bottleneck in high-resolution video generation. Training-free sparse attention r…

cs.CV2026

Chorus II: Cross-Request Sparsity Reuse for Efficient Image-to-Video Generation

Hao Liu, Chenghuan Huang, Xing Cai +5

Serving diffusion models for image-to-video generation is computationally expensive, posing significant challenges for large-scale deployment. Real I2V workloads often contain simi…

cs.CV2026

Beyond Few-Step Inference: Accelerating Video Diffusion Transformer Model Serving with Inter-Request Caching Reuse

Hao Liu, Ye Huang, Chenghuan Huang +5

Video Diffusion Transformer (DiT) models are a dominant approach for high-quality video generation but suffer from high inference cost due to iterative denoising. Existing caching…

cs.LG2026

Pretrain Value, Not Reward: Decoupled Value Policy Optimization

Chenghua Huang, Lu Wang, Fangkai Yang +6

In this paper, we explore how directly pretraining a value model simplifies and stabilizes reinforcement learning from human feedback (RLHF). In reinforcement learning, value estim…

cs.CL2025

Self-Evolved Reward Learning for LLMs

Chenghua Huang, Zhizhen Fan, Lu Wang +7

Reinforcement Learning from Human Feedback (RLHF) is a crucial technique for aligning language models with human preferences, playing a pivotal role in the success of conversationa…

cs.CL2025

Distill Not Only Data but Also Rewards: Can Smaller Language Models Surpass Larger Ones?

Yudi Zhang, Lu Wang, Meng Fang +8

Distilling large language models (LLMs) typically involves transferring the teacher model's responses through supervised fine-tuning (SFT). However, this approach neglects the pote…