Yuchen Ma 马羽宸

I am a Ph.D. student in Computer Science at LMU Munich and the Munich Center for Machine Learning (MCML), supervised by Prof. Stefan Feuerriegel. Before that, I received my M.Sc. in Mathematics & Computer Vision from Heidelberg University.

I interned at Microsoft Research in Seattle from June to September 2026, working with Swadheen Shukla and Michel Galley. During my internship, I developed EAIR, a framework for diagnosing and mitigating interaction risks in enterprise LLM multi-agent systems.

My research interests are in agentic AI, large language models, and causal inference. Specifically, I build causal foundation models, develop scalable methods for personalizing LLMs, study how LLM agents can distill verified, reusable skills, and explore how multiple agents interact and collaborate. I'm happy to connect and discuss potential collaborations.

Yuchen Ma | 马羽宸

News

Oct 2026 New first-author preprint: EAIR: Interaction Risks in Enterprise LLM Multi-Agent Systems, developed during my internship at Microsoft Research. Toolkit.
Sep 2026 Two first-author papers accepted at NeurIPS 2026. See you in Sydney!
Jun 2026 Joining Microsoft Research as a research intern this summer. I'll be in Seattle this summer, happy to meet for a coffee chat!
May 2026 Research stay at University of Notre Dame (May – Jun 2026).
May 2026 One first-author paper accepted at KDD 2026. See you in Jeju!
May 2026 One first-author paper accepted at ICML 2026. See you in Seoul!
Apr 2026 Organizing the RelSciFM @ KDD 2026 workshop on Reliable Scientific Foundation Models.
Jan 2026 One first-author paper accepted at ICLR 2026. See you in Rio de Janeiro!
Jan 2026 Released the CausalFM toolkit — docs.
Sep 2025 One first-author paper accepted at NeurIPS 2025. See you in San Diego!
May 2025 One first-author paper accepted at KDD 2025. See you in Toronto!
Sep 2024 One first-author paper accepted at NeurIPS 2024. See you in Vancouver!

Selected Projects

Agent Harness

Agent Harness

We study how LLM agents can distill their own experience into reusable, verified skills that improve capabilities without retraining.

Highlights:

Other related works: MemoHarness: Agent Harnesses That Learn from Experience (arXiv'26), NARRA-Gym for Evaluating Interactive Narrative Agents (NeurIPS'26)

Multi-Agent Systems

Multi-Agent Systems

We study how LLM agents interact and collaborate in multi-agent and enterprise settings, and build tools to diagnose and mitigate the risks that emerge from their interactions.

Highlights:

Other related works: Emergent Risks in Generative Multi-Agent Systems (Nature MI'26)

Applicable and Reliable LLMs

Applicable and Reliable LLMs

We develop scalable approaches for adapting LLMs to individual users' preferences and latent constraints.

Highlights:

Other related works: ProbeLLM (ICML'26), Treatment Effects under Text Confounding (NeurIPS'25)

Causal graphs for causal inference settings Multi-outcome distribution of treatments
Diffusion-based estimation of multi-outcome distributions

Causal Foundation Models

We build foundation models that perform causal inference on new datasets in a training-free way via in-context learning.

Highlights:

Other related works: DiffPO (NeurIPS'24), Multi-Outcome Distributions of Treatments (KDD'25), Treatment Effects under Text Confounding (NeurIPS'25), Counterfactual Fairness (CLeaR'26)

Recent Preprints & Publications

/ view by year / view by topic

* = Equal contribution. The full publication list can be found on my Google Scholar profile.

Education

Ph.D.2022 – Present
LMU Munich, Germany
Ph.D. student in Computer Science
M.Sc.2019 – 2022
Heidelberg University, Germany
M.Sc. in Mathematics & Computer Vision
Advisor: Prof. Zeynep Akata
B.Sc.2015 – 2019
Shandong University, China
B.Sc. in Mathematics
Advisor: Prof. Guanghui Wang

Experience

Research InternJun – Sep 2026
Microsoft Research, Seattle, U.S.
Advisors: Swadheen Shukla, Michel Galley
Developed EAIR for diagnosing and mitigating enterprise multi-agent interaction risks, with a testable risk taxonomy, executable micro-worlds, and trace-guided workflow repair.
Visiting ResearcherMay – Jun 2026
University of Notre Dame, South Bend, IN, U.S.
Advisor: Prof. Xiangliang Zhang
Research AssistantJun 2021 – Feb 2022
Max Planck Institute, Germany
Advisor: Prof. Zeynep Akata