Shengen Yan
10 papers · 2024–2025 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+2 more ↓ Show less ↑
π Interdisciplinary Bridge π§ Keyword Pioneer π Cross-Pollinator (15) π Conference Polyglot (6) πΊοΈ Taxonomy Completionist (13)
π
Century Club
(10)
β‘
Prolific Year
(6)
Conferences
ICCV (3)
ICLR (2)
NIPS (2)
CVPR (1)
ECCV (1)
ICML (1)
Top co-authors
Keywords
diffusion transformer
(3)
vision-language model
(2)
efficient computing
(2)
video generation
(2)
model compression
(2)
text-to-image generation
(1)
computational efficiency
(1)
feature similarity
(1)
diffusion model
(1)
window attention
(1)
key-value cache
(1)
latency optimization
(1)
memory management
(1)
token merging
(1)
token reduction
(1)
multimodal generation
(1)
latent frame rate
(1)
motion frequency
(1)
rotary positional embedding
(1)
transformer model
(1)
Papers
FrameFusion: Combining Similarity and Importance for Video Token Reduction on Large Vision Language Models
ICCV 2025
Linear Combination of Saved Checkpoints Makes Consistency and Diffusion Models Better
ICLR 2025
ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation
ICLR 2025
MBQ: Modality-Balanced Quantization for Large Vision-Language Models
CVPR 2025
DLFR-Gen: Diffusion-based Video Generation with Dynamic Latent Frame Rate
ICCV 2025
DiTFastAttnV2: Head-wise Attention Compression for Multi-Modality Diffusion Transformers
ICCV 2025
Evaluating Quantized Large Language Models
ICML 2024
ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction
NIPS 2024
MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization
ECCV 2024
DiTFastAttn: Attention Compression for Diffusion Transformer Models
NIPS 2024