Scholar
Feng Yao
Google Scholar ID: huCn_LwAAAAJ
University of California, San Diego
Citations & Impact
Citations
327
H-index
10
i10-index
10
Publications
20
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- [{'PaperTitle': 'FlashRL: 8Bit Rollouts, Full Power RL', 'Authors': 'Liyuan Liu*, Feng Yao* et al.', 'PublicationStatus': 'Preprint', 'PublicationYear': '2025'}, {'PaperTitle': 'Your Efficient RL Framework Secretly Brings You Off-Policy RL Training', 'Authors': 'Feng Yao*, Liyuan Liu* et al.', 'PublicationStatus': 'Preprint', 'PublicationYear': '2025'}, {'PaperTitle': 'DenseMixer: Improving MoE Post-Training with Precise Router Gradients', 'Authors': 'Feng Yao et al.', 'PublicationStatus': 'Preprint', 'PublicationYear': '2025'}, {'PaperTitle': 'Training Language Models to Generate Quality Code with Program Analysis Feedback', 'Authors': 'Feng Yao*, Zilong Wang* et al.', 'PublicationStatus': 'NeurIPS 2025', 'PublicationYear': '2025'}, {'PaperTitle': 'Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective', 'Authors': 'Zhoujun Cheng*, Shibo Hao* et al.', 'PublicationStatus': 'Preprint', 'PublicationYear': '2025'}, {'PaperTitle': 'Data Contamination Can Cross Language Barriers', 'Authors': 'Feng Yao*, Yufan Zhuang* et al.', 'PublicationStatus': 'EMNLP 2024', 'PublicationYear': '2024'}, {'PaperTitle': 'Configurable Foundation Models: Building LLMs from a Modular Perspective', 'Authors': 'Chaojun Xiao, Zhengyan Zhang et al.', 'PublicationStatus': 'Preprint', 'PublicationYear': '2024'}]
Background
- Research Interests: Intersection of Natural Language Processing and Deep Learning. Recently focused on training Mixture-of-Experts (MoE) and improving the efficiency of large-scale Reinforcement Learning.
Miscellany
- Invited Talks:
- - Applied Compute (ex-OpenAI startup), November 10, 2025
- - Tencent Hunyuan Team, October 23, 2025
- - H2Lab@UW, October 22, 2025
- - SeaAI Lab, September 03, 2025
- - MiniMax, September 01, 2025
- - QingkeAI, August 29, 2025
- - TsinghuaNLP & ModelBest, August 27, 2025
- - Kuaishou Klear Team, August 19, 2025
- - Cohere Labs, July 14, 2025
- - Qwen Team, July 02, 2025
