|
Yilin Guan I am a master student at Johns Hopkins University, where I am fortunate to be advised by Prof. Daniel Khashabi. I also work closely with Prof. Tianlong Chen and Prof. Kaidi Xu. Previously, I obtained my B.S. of Computer Science from University of Chinese Academy of Sciences with honor. My research interests include efficient agent system and efficient machine learning. Feel free to contact me if you want to discuss or collaborate! |
|
News 📢
|
Publications 📜 |
|
B3-PWL: GPU-Batched Branch-and-Bound for Piecewise-Linear Optimization with SOS2 Constraints
Yilin Guan, Shuqing Luo, Pingzhi Li, Tianlong Chen, Kaidi Xu arXiv preprint, 2026 arXiv A GPU-centric branch-and-bound framework for piecewise-linear optimization with SOS2 constraints: batches of LP relaxations are solved concurrently with a first-order primal-dual solver and a batched block-tiled sparse kernel, with a unified feasibility-search module for fast incumbents — 9.25x geometric-mean speedup over NVIDIA cuOpt on 43 PWL-MIP instances. |
|
AsyncSpade: Efficient Test-Time Scaling with Asynchronous Sparse Decoding
Shuqing Luo*, Yilin Guan*, Pingzhi Li, Hanrui Wang, Tianlong Chen * Equal contribution ICML, 2026 arXiv Asynchronous framework for efficient test-time scaling: light-weight temporal-regressive query prediction and disaggregated KV-cache filtering overlapped with inference. |
|
Dynamic Speculative Agent Planning
Yilin Guan, Wenyue Hua, Qingfeng Lan, Fei Sun, Dujian Ding, Devang Acharya, Chi Wang, William Yang Wang ICLR, 2026 arXiv / code A dynamic speculative planning framework for LLM-based agents that accelerates multi-step reasoning by adaptively speculating future actions. |
|
Template from Jon Barron. |
