4 citations · 12 across the 10 of their papers we have counts for
7 papers · 1 filter
OmniSparse: Training-Aware Fine-Grained Sparse Attention for Long-Video MLLMs
Feng Chen, Yefei He, Shaoxuan He +9
Existing sparse attention methods primarily target inference-time acceleration by selecting critical tokens under predefined sparsity patterns. However, they often fail to bridge t…
ZipVL: Efficient Large Vision-Language Models with Dynamic Token Sparsification
Yefei He, Feng Chen, Jing Liu +4
The efficiency of large vision-language models (LVLMs) is constrained by the computational bottleneck of the attention mechanism during the prefill phase and the memory bottleneck…
DragAnything: Motion Control for Anything using Entity Representation
Weijia Wu, Zhuang Li, Yuchao Gu +7
We introduce DragAnything, which utilizes a entity representation to achieve motion control for any object in controllable video generation. Comparison to existing motion control m…
Towards Accurate Post-training Quantization for Reparameterized Models
Luoming Zhang, Yefei He, Wen Fei +4
Model reparameterization is a widely accepted technique for improving inference speed without compromising performance. However, current Post-training Quantization (PTQ) methods of…
DSText V2: A Comprehensive Video Text Spotting Dataset for Dense and Small Text
Weijia Wu, Yiming Zhang, Yefei He +4
Recently, video text detection, tracking, and recognition in natural scenes are becoming very popular in the computer vision community. However, most existing algorithms and benchm…
Paragraph-to-Image Generation with Information-Enriched Diffusion Model
Weijia Wu, Zhuang Li, Yefei He +6
Text-to-image (T2I) models have recently experienced rapid development, achieving astonishing performance in terms of fidelity and textual alignment capabilities. However, given a…
EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models
Yefei He, Jing Liu, Weijia Wu +2
Diffusion models have demonstrated remarkable capabilities in image synthesis and related generative tasks. Nevertheless, their practicality for real-world applications is constrai…