Scholar
Tianshi Xu
Google Scholar ID: YS2vV5IAAAAJ
PhD @ Peking University
Citations & Impact
Citations
45
H-index
4
i10-index
2
Publications
13
Co-authors
15
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Paper “CryptoMoE: Privacy-Preserving and Scalable Mixture of Experts Inference via Balanced Expert Routing” accepted by NeurIPS 2025 (CCF A)
- Paper “EfficientNav: Towards On-Device Object-Goal Navigation with Navigation Map Caching and Retrieval” accepted by NeurIPS 2025 (CCF A)
- Collaboration paper “Ironman: Accelerating Oblivious Transfer Extension for Privacy-Preserving AI with Near-Memory Processing” accepted by MICRO'25 (CCF A)
- Paper “Breaking the Layer Barrier: Remodeling Private Transformer Inference with Hybrid CKKS and MPC” accepted by USENIX Security 2025 (CCF A)
- Collaboration paper “Swift: Fast Secure Neural Network Inference with Fully Homomorphic Encryption” accepted by IEEE TIFS (CCF A)
- Multiple publications at top-tier venues including NeurIPS, USENIX Security, DAC, MICRO, ICCAD
- NeurIPS Scholar Award, October 2025
- Boya Scholarship, Peking University, November 2024
- Excellent Research Award, Peking University, September 2024
- National Scholarship (Ministry of Education), April 2023 (ranked 1st/189)
- Merit Student in Beijing, March 2023 (ranked 2nd/189)
- SAMSUNG Scholarship, June 2022 (top 11 among 3815 undergraduates)
- Professional Membership of ACM, October 2024
Background
- Second-year Ph.D. student in Artificial Intelligence at Peking University, advised by Prof. Meng Li
- Currently a research intern at Microsoft Research (Asia)
- Exploring mathematical reasoning capabilities of large language models (LLMs)
- Previously focused on efficient, privacy-preserving deep learning algorithms/systems and applied cryptography
- Aims to contribute solid, long-term work to the field
