Han Cai

papers

764

total citations

papers (15)

APQ: Joint Search for Network Architecture, Pruning and Quantization Policy

CVPR 2020arXiv

205

citations

Memory-efficient Patch-based Inference for Tiny Deep Learning

NEURIPS 2021arXiv

183

citations

DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models

CVPR 2024arXiv

104

citations

Lite Pose: Efficient Architecture Design for 2D Human Pose Estimation

CVPR 2022arXiv

citations

SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation

ICCV 2025arXiv

citations

Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation

NEURIPS 2025arXiv

citations

DC-AE 1.5: Accelerating Diffusion Model Convergence with Structured Latent Space

ICCV 2025arXiv

citations

COAT: Compressing Optimizer states and Activations for Memory-Efficient FP8 Training

ICLR 2025arXiv

citations

Win Fast or Lose Slow: Balancing Speed and Accuracy in Latency-Sensitive Decisions of LLMs

NEURIPS 2025arXiv

citations

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

ICCV 2025arXiv

citations

TinyTL: Reduce Memory, Not Parameters for Efficient On-Device Learning

NEURIPS 2020

citations

EfficientViT: Lightweight Multi-Scale Attention for High-Resolution Dense Prediction

ICCV 2023

citations

Han Cai

papers (15)

APQ: Joint Search for Network Architecture, Pruning and Quantization Policy

Memory-efficient Patch-based Inference for Tiny Deep Learning

DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models

Lite Pose: Efficient Architecture Design for 2D Human Pose Estimation

SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation

Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation

DC-AE 1.5: Accelerating Diffusion Model Convergence with Structured Latent Space

COAT: Compressing Optimizer states and Activations for Memory-Efficient FP8 Training

Condition-Aware Neural Network for Controlled Image Generation

Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search

Scaling Vision Pre-Training to 4K Resolution

Win Fast or Lose Slow: Balancing Speed and Accuracy in Latency-Sensitive Decisions of LLMs

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

TinyTL: Reduce Memory, Not Parameters for Efficient On-Device Learning

EfficientViT: Lightweight Multi-Scale Attention for High-Resolution Dense Prediction

papers (15)

APQ: Joint Search for Network Architecture, Pruning and Quantization Policy

Memory-efficient Patch-based Inference for Tiny Deep Learning

DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models

Lite Pose: Efficient Architecture Design for 2D Human Pose Estimation

SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation

Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation

DC-AE 1.5: Accelerating Diffusion Model Convergence with Structured Latent Space

COAT: Compressing Optimizer states and Activations for Memory-Efficient FP8 Training

Condition-Aware Neural Network for Controlled Image Generation

Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search

Scaling Vision Pre-Training to 4K Resolution

Win Fast or Lose Slow: Balancing Speed and Accuracy in Latency-Sensitive Decisions of LLMs

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

TinyTL: Reduce Memory, Not Parameters for Efficient On-Device Learning

EfficientViT: Lightweight Multi-Scale Attention for High-Resolution Dense Prediction