Picture for Huan Yang

Huan Yang

Depatment of Gastroenterology, Second Affiliated Hospital, Army Medical University

KVShare: Semantic-Aware Key-Value Cache Sharing for Efficient Large Language Model Inference

Add code
Mar 17, 2025
Viaarxiv icon

Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers

Add code
Mar 14, 2025
Viaarxiv icon

Accelerating Video Diffusion Models via Distribution Matching

Add code
Dec 08, 2024
Figure 1 for Accelerating Video Diffusion Models via Distribution Matching
Figure 2 for Accelerating Video Diffusion Models via Distribution Matching
Figure 3 for Accelerating Video Diffusion Models via Distribution Matching
Figure 4 for Accelerating Video Diffusion Models via Distribution Matching
Viaarxiv icon

Long Video Diffusion Generation with Segmented Cross-Attention and Content-Rich Video Data Curation

Add code
Dec 02, 2024
Figure 1 for Long Video Diffusion Generation with Segmented Cross-Attention and Content-Rich Video Data Curation
Figure 2 for Long Video Diffusion Generation with Segmented Cross-Attention and Content-Rich Video Data Curation
Figure 3 for Long Video Diffusion Generation with Segmented Cross-Attention and Content-Rich Video Data Curation
Figure 4 for Long Video Diffusion Generation with Segmented Cross-Attention and Content-Rich Video Data Curation
Viaarxiv icon

VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation

Add code
Dec 01, 2024
Figure 1 for VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation
Figure 2 for VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation
Figure 3 for VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation
Figure 4 for VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation
Viaarxiv icon

Fleximo: Towards Flexible Text-to-Human Motion Video Generation

Add code
Nov 29, 2024
Viaarxiv icon

Improving Multi-Subject Consistency in Open-Domain Image Generation with Isolation and Reposition Attention

Add code
Nov 28, 2024
Viaarxiv icon

Allegro: Open the Black Box of Commercial-Level Video Generation Model

Add code
Oct 20, 2024
Figure 1 for Allegro: Open the Black Box of Commercial-Level Video Generation Model
Figure 2 for Allegro: Open the Black Box of Commercial-Level Video Generation Model
Figure 3 for Allegro: Open the Black Box of Commercial-Level Video Generation Model
Figure 4 for Allegro: Open the Black Box of Commercial-Level Video Generation Model
Viaarxiv icon

Self-Supervised Learning of Deviation in Latent Representation for Co-speech Gesture Video Generation

Add code
Sep 26, 2024
Figure 1 for Self-Supervised Learning of Deviation in Latent Representation for Co-speech Gesture Video Generation
Figure 2 for Self-Supervised Learning of Deviation in Latent Representation for Co-speech Gesture Video Generation
Figure 3 for Self-Supervised Learning of Deviation in Latent Representation for Co-speech Gesture Video Generation
Figure 4 for Self-Supervised Learning of Deviation in Latent Representation for Co-speech Gesture Video Generation
Viaarxiv icon

A First Look At Efficient And Secure On-Device LLM Inference Against KV Leakage

Add code
Sep 06, 2024
Figure 1 for A First Look At Efficient And Secure On-Device LLM Inference Against KV Leakage
Figure 2 for A First Look At Efficient And Secure On-Device LLM Inference Against KV Leakage
Figure 3 for A First Look At Efficient And Secure On-Device LLM Inference Against KV Leakage
Figure 4 for A First Look At Efficient And Secure On-Device LLM Inference Against KV Leakage
Viaarxiv icon