Scholar
Wei-Lin Chiang
Google Scholar ID: 87nZphcAAAAJ
UC Berkeley
Citations & Impact
Citations
14,159
H-index
19
i10-index
25
Publications
20
Co-authors
0
Publications
Cited
0
Resume (English only)
Academic Achievements
- Published "Judging LLM-as-a-judge with MT-Bench and Chatbot Arena" at NeurIPS 2023
- Published two papers at ICLR 2024: "LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset" and "Llm-assisted code cleaning for training accurate code generators"
- Released technical report on Chatbot Arena (arXiv preprint)
- Proposed automated LLM evaluation benchmarks such as MT-Bench and Arena-Hard
- Vicuna has been downloaded over 8 million times with 1000+ citations
- FastChat has garnered over 30K GitHub stars and 200+ contributors
- Hosted a Kaggle competition for human preference prediction (May 2024)
- Cluster-GCN is widely integrated into platforms like DGL and PyTorch Geometric
Research Experience
- Conducting AI evaluation research at SkyLab
- Leading or contributing to multiple AI evaluation and LLM-related projects, including Chatbot Arena, Multimodal Arena, RedTeam Arena, WebDev Arena, Arena Hard, and BenchBuilder
- Developed FastChat, a multi-model serving framework powering Chatbot Arena
- Contributed to Vicuna, a high-quality LLM chatbot
- Worked on SkyPilot, an intercloud system for AI and batch jobs
- Developed Cluster-GCN for scalable training of large graph neural networks
Co-authors