Chaojian Li

11 papers · 2020–2026 · 6 conferences · across top CS/AI conferences

Achievements

🐝 Cross-Pollinator (13) 🌍 Conference Polyglot (5) 🏃 Academic Marathon (5) 🌈 Renaissance Researcher (5) 🌉 Interdisciplinary Bridge 🗺️ Taxonomy Completionist (16) 🧭 Keyword Pioneer 🐣 Hot Topic Early Bird 💎 Century Club (10)

Conferences

NIPS (4) ECCV (2) ICLR (2) AAAI (1) CVPR (1) EACL (1)

Top co-authors

Yingyan Lin (8) Yonggan Fu (6) Haoran You (6) Zhangyang Wang (4) Yongan Zhang (3) Yue Wang (3) Zhifan Ye (2) Sixu Li (2) Xiaohan Chen (2) Sicheng Li (2)

Research topics

Keywords

efficient computing (3) model compression (2) vision transformer (1) efficient training (1) adversarial robustness (1) neural rendering (1) computational efficiency (1) machine teaching (1) adaptive computation (1) diffusion model (1) model training (1) 3d gaussian splatting (1) novel view synthesis (1) best-of-n sampling (1) personalized education (1) low-precision training (1) beam search (1) gradient quantization (1) iterative machine teaching (1) process reward model (1)

Papers

Think Hard Only When Needed: A Hybrid Best-of-N and Beam Search for Efficient Test-Time Compute EACL 2026

Early-Bird Diffusion: Investigating and Leveraging Timestep-Aware Early-Bird Tickets in Diffusion Models for Efficient Training CVPR 2025

Omni-Recon: Harnessing Image-based Rendering for General-Purpose Neural Radiance Fields ECCV 2024

3D Gaussian Rendering Can Be Sparser: Efficient Rendering via Learned Fragment Pruning NIPS 2024

MIA-Former: Efficient and Robust Vision Transformers via Multi-Grained Input-Adaptation AAAI 2022

HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark ICLR 2021

Locality Sensitive Teaching NIPS 2021

FracTrain: Fractionally Squeezing Bit Savings Both Temporally and Spatially for Efficient DNN Training NIPS 2020

ShiftAddNet: A Hardware-Inspired Deep Network NIPS 2020

HALO: Hardware-Aware Learning to Optimize ECCV 2020

Drawing Early-Bird Tickets: Toward More Efficient Training of Deep Networks ICLR 2020