close
Portrait of Kexin Tan

Kexin Tan

譚可欣 · You may also call me KT

GitHub Email CV

About

Currently an M.Eng. student in TMI, School of Engineering, The University of Tokyo.

My earlier academic training includes a B.S. in Mathematical Economics from Fudan University and a brief period in Georgia Tech's OMSCS program.

Previously, I spent three years working on data-driven growth strategy at a FinTech SaaS company, where I became familiar with and fond of the pace and ambition of startup environments.

Academic Interests

I am interested in how general intelligent systems, from humans to language models, reason; and how such reasoning can be evaluated with care rather than reduced to surface-level scores.

My recent academic work has touched on LLM evaluation, logic-and-NLP benchmarks, and AI systems for scholarly workflows.

Publications

* denotes co-first authors. † denotes corresponding author.

OpenNovelty: An LLM-Powered Agentic System for Verifiable Scholarly Novelty Assessment Ming Zhang*†, Kexin Tan*, Yueyuan Huang*, Yujiong Shen, Chunchun Ma, Li Ju, Xinran Zhang, Yuhui Wang, Wenqing Jing, Jingyi Deng, et al. Paper · Code · Report Website
LLMEval-Fair: A Large-Scale Longitudinal Study on Robust and Fair Evaluation of Large Language Models Ming Zhang*†, Yujiong Shen*, Jingyi Deng*, Yuhui Wang*, Huayu Sha, Kexin Tan, Qiyuan Peng, Yue Zhang, Junzhe Wang, et al. Paper · Code
Can Deep Research Agents Retrieve and Organize? Evaluating the Synthesis Gap with Expert Taxonomies Ming Zhang*†, Jiabao Zhuang*, Wenqing Jing*, Kexin Tan*, Ziyu Kong, Jingyi Deng, Yujiong Shen, Yuhang Zhao, Ning Luo, et al. Paper
WisPaper: Your AI Scholar Search Engine Li Ju*, Jun Zhao*, Mingxu Chai, Ziyu Shen, Xiangyang Wang, Yage Geng, Chunchun Ma, …, Ming Zhang, Kexin Tan, Jujie Ye, et al. Paper

Selected Experience

Research Assistant Intern, NLP Lab, Fudan University Working on language-model evaluation and interdisciplinary benchmark research at the intersection of logic and NLP.
Growth Lead, Data & Strategy, 1Token Inc. Led data-driven growth for an asset-management infrastructure provider serving 50+ institutions worldwide. Built global institution databases, analyzed lead pipelines and funding networks, designed product-led-growth trials, automated cross-team workflows, and improved online growth by 150% through website and SEO optimization.
Research and Strategy Internships Earlier work across operations strategy, ESG data analysis, economic research assistance, and green-finance literature review.

Education

The University of Tokyo
Georgia Institute of Technology Withdrew from OMSCS due to scheduling conflicts with work.
Fudan University
Chengdu Foreign Languages School Graduated with First Place.

Contact

Find me on GitHub, email me at bluellatan@gmail.com, or download my CV.