Lingfeng Shen

MAI SuperIntelligence Team

Mountain View, CA

I am a Member of Technical Staff on the MAI SuperIntelligence Team, where I work on agentic foundation models. Before coming to MAI, I was a research scientist at ByteDance Seed, focusing on the reasoning, planning and function-call capacities of LLMs. Apart from research, I consider myself a professional trader in event-driven trading and risk hedging.

During my time at Johns Hopkins, I worked with Daniel Khashabi and Philipp Koehn. Besides, I was fortunate to spend my last year of undergraduate at Tencent AI Lab, mentored by Haiyun Jiang, Lemao Liu and Shuming Shi.

If you are interested in chatting about research with me, I am always open to discussions and academic collaborations. Feel free to reach out to me at lshen30[at]jh.edu or lingfengshen[at]microsoft.com.

Academic Services

Reviewer: ICLR 2024, NeurIPS 2023/2024, ICML 2023, ACL 2023, AAAI 2022/2023/2024, EMNLP 2022/2023, ARR 2022/2023.

Research Interests

Emergent behaviors of foundation models, e.g. in-context learning — paper
Data-centric design in building foundation models — paper
Adversarial robustness — paper, paper
Backdoor robustness — paper
Evaluation of generation models — paper, paper

Beyond Research

Motorsports ignite my passion, particularly Formula 1 — I am ardently devoted to the Mercedes-AMG Petronas F1 Team.

The NBA also has a special place in my heart; I used to be a big fan of the Houston Rockets.

Off the screen, you will often find me karting with friends or playing video games.