7 papers · 1 filter
LoopREG: Representation Supervision Enables Test-Time Recurrent Depth in Diffusion Transformers
Zipeng Wang, Anping Xiong, Jing Zhang +17
Reusing Transformer blocks offers a way to increase diffusion-model computation at test time, but naive recurrence can degrade generation beyond its training depth. We introduce Lo…
Post-Training Pruning for Diffusion Transformers
Chengzhi Hu, Xuewen Liu, Jing Zhang +3
Diffusion Transformers (DiTs) have demonstrated impressive performance in image generation but suffer from substantial computational overhead and resource consumption. Post-trainin…
OSAQ: Outlier Self-Absorption for Accurate Low-bit LLM Quantization
Zhikai Li, Zhen Dong, Xuewen Liu +2
Large Language Models (LLMs) have demonstrated remarkable capabilities. However, their massive parameter scale leads to significant resource consumption and latency during inferenc…
Arena as Offline Reward: Efficient Fine-Grained Preference Optimization for Diffusion Models
Zhikai Li, Yue Zhao, Edward Zhongwei Zhang +4
Reinforcement learning from human feedback (RLHF) effectively promotes preference alignment of text-to-image (T2I) diffusion models. To improve computational efficiency, direct pre…
Sparsity Induction for Accurate Post-Training Pruning of Large Language Models
Minhao Jiang, Zhikai Li, Xuewen Liu +3
Large language models have demonstrated capabilities in text generation, while their increasing parameter scales present challenges in computational and memory efficiency. Post-tra…
Efficient-SAM2: Accelerating SAM2 with Object-Aware Visual Encoding and Memory Retrieval
Jing Zhang, Zhikai Li, Xuewen Liu +1
Segment Anything Model 2 (SAM2) shows excellent performance in video object segmentation tasks; however, the heavy computational burden hinders its application in real-time video p…
PTQ4ARVG: Post-Training Quantization for AutoRegressive Visual Generation Models
Xuewen Liu, Zhikai Li, Jing Zhang +2
AutoRegressive Visual Generation (ARVG) models retain an architecture compatible with language models, while achieving performance comparable to diffusion-based models. Quantizatio…