Tianle Li
Member of technical staff at xAI (Science of RL team); lead author of Arena-Hard and a core builder of LMArena while at UC Berkeley.
- Role
- research-scientist
- Currently at
- xAI
- twitter.com/CodingWithTim
- GitHub
- github.com/CodingWithTim
- Scholar
- scholar.google.com/citations
- Papers
- 11
Cite
Notes
Only stored in your browser.
Authored papers
11Prompt-to-Leaderboard
arXiv 2025
Search Arena: Analyzing Search-Augmented LLMs
arXiv 2025
Differentiable Evolutionary Reinforcement Learning
arXiv 2025
From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline
preprint
Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
ICML
MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark
NeurIPS
Style Control: Does Style Matter? A Closer Look at Chatbot Arena
blog
Long-context LLMs Struggle with Long In-context Learning
arXiv 2024
How to Evaluate Reward Models for RLHF
arXiv 2024
SWAG: Storytelling With Action Guidance
arXiv 2024
ImagenHub: Standardizing the evaluation of conditional image generation models
arXiv 2023
Affiliations
Frequent co-authors
10from 11 papers
Wei-Lin Chiang
co-founder / President
Evan Frick
research-scientist
Ion Stoica
professor / co-founder
Anastasios N. Angelopoulos
Banghua Zhu
professor
Joseph E. Gonzalez
Wenhu Chen
professor
Anastasios N. Angelopoulos
CEO
Connor Chen
Ge Zhang
researcher