Haocheng Xi

8 papers · 2023–2025 · 4 conferences · across top CS/AI conferences

Achievements

🌍 Conference Polyglot (4) 🌉 Interdisciplinary Bridge 🐝 Cross-Pollinator (14) 👥 Mega-Team (25) ⚡ Prolific Year (6)

Conferences

ICML (5) CVPR (1) ICLR (1) NIPS (1)

Top co-authors

Jianfei Chen (6) Jun Zhu (4) Song Han (3) Kurt Keutzer (3) Xiuyu Li (2) Yuxiang Chen (2) Yao Lu (2) Jintao Zhang (2) Dacheng Li (2) Han Cai (2)

Keywords

efficient computing (2) neural network training (1) low-precision training (1) gradient quantization (1) token compression (1) visual language model (1) integer arithmetic (1) model quantization (1) spatial resolution (1)

Papers

SpargeAttention: Accurate and Training-free Sparse Attention Accelerating Any Model Inference ICML 2025

NVILA: Efficient Frontier Visual Language Models CVPR 2025

COAT: Compressing Optimizer states and Activations for Memory-Efficient FP8 Training ICLR 2025

QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache ICML 2025

Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity ICML 2025

Oscillation-Reduced MXFP4 Training for Vision Transformers ICML 2025

Jetfire: Efficient and Accurate Transformer Pretraining with INT8 Data Flow and Per-Block Quantization ICML 2024

Training Transformers with 4-bit Integers NIPS 2023