Scholar
Yifei Wang
Google Scholar ID: -CLy6YsAAAAJ
MIT
Citations & Impact
Citations
1,733
H-index
23
i10-index
34
Publications
20
Co-authors
21
list available
Publications
Resume (English only)
Academic Achievements
- Four papers accepted at NeurIPS 2025, including RL-driven graph reasoning for LLMs (G1), signed graph propagation, hierarchical diffusion LMs, and ranking-based LLM decoding.
- Three papers accepted at ICML 2025, including an oral presentation (top 1%) introducing Contrastive Sparse Representations (CSR) for highly efficient embedding compression.
- Best Paper Runner-Up Award at ICLR 2025 Workshop on Reasoning and Planning for LLMs for the paper 'When More is Less: Understanding Chain-of-Thought Length in LLMs'.
- ICML 2025 paper 'On the Emergence of Position Bias in Transformers' featured by MIT News for theoretically identifying root causes of position bias in Transformers.
- Serves as Area Chair for ICLR and reviewer for ICML, NeurIPS, JMLR, PNAS.
- Recent work has received 4 best paper awards and has been featured by MIT News, CSAIL News, and Anthropic.
