Andrew, Jingkun Ma

Andrew, Jingkun Ma

Ph.D. Student

University of Macau

Research Interests

Trustworthy and reasoning-driven AI
Formal theorem proving
Autonomous agents
Cultural alignment
AI for healthcare, education, and personalized decision support

Home

I am Andrew, Jingkun Ma, a second-year Ph.D. student in Computer Science at the University of Macau and a member of the NLP2CT Lab, supervised by Prof. Derek F. Wong. I am deeply grateful to Prof. Derek F. Wong for his invaluable supervision, support, and academic guidance throughout my doctoral journey. I would also like to express my sincere gratitude to Runzhe Zhan for his generous mentorship, insightful discussions, and continuous encouragement, all of which have played an important role in shaping my research interests and development. In addition, I have greatly benefited from the valuable insights and guidance of Dr. Hou Pong Chan from Alibaba DAMO Academy. My research focuses on building trustworthy and reasoning-driven AI systems. I am particularly interested in formal theorem proving, agentic systems, and culturally aligned AI, with applications in healthcare, education, and personalized decision support. Beyond research, I am the founder of the Artificial Intelligence and Data Science Society (AIDSS) at the University of Macau, where I serve as Chief Academic Mentor.

Research Group Overview

I currently mentor a growing group of undergraduate and master's students working on formal theorem proving, compositional reasoning, agents, medical AI, multi-modal LLMs, Math + AI, and cultural alignment.

Recent mentee outcomes include 2026 master's admissions to Johns Hopkins University, HKUST, CUHK, and the University of Macau.

Highlighted students:

  • Yuchao Wang: Formal theorem proving; co-authored FIPO-Prover, accepted to ICML 2026 Workshop AI4Math with a Spotlight
  • Yang Li: Multi-modal LLM and Math + AI; co-authored VisAidMath, which received the Best Paper Award at NeurIPS 2025 Workshop VLM4RWD
  • Rongsheng Zhang: Math + AI, admitted to Johns Hopkins University M.S. (2026)
  • Lingyi Ou: Medical AI, admitted to HKUST M.S. (2026)
  • Xingchen Chen: Medical AI, admitted to University of Macau M.S. (2026)
  • Zhexuan Zhang: Cultural Alignment, admitted to CUHK M.S. (2026)

More details can be found on the Research Group page.

Selected Publications

View All ->
VisAidMath: Benchmarking Visual-Aided Mathematical Reasoning

VisAidMath: Benchmarking Visual-Aided Mathematical Reasoning

Jingkun Ma, Runzhe Zhan, Yang Li, Di Sun, Hou Pong Chan, Lidia S. Chao, Derek F. Wong

ACL 2026NeurIPS 2025 Workshop@VLM4RWD (Best Paper Award)

A benchmark for visual-aided mathematical reasoning, accepted by ACL 2026 and also recognized with the Best Paper Award at the 1st VLM4RWD Workshop at NeurIPS 2025.

FIPO-Prover: Formalization-Oriented Informal Proof Optimization for Efficient Formal Theorem Proving

FIPO-Prover: Formalization-Oriented Informal Proof Optimization for Efficient Formal Theorem Proving

Jingkun Ma*, Yuchao Wang*, Yujia Huo, Derek F. Wong

ICML 2026 Workshop@AI4Math (Spotlight Award)

FIPO-Prover bridges the gap between informal reasoning and formal verification by optimizing an intermediate proof representation that is diagnosable, repairable, and guided by verifier feedback.

Idioms Understood, Yet Translated Literally: Diagnosing Literal Translation Bias in Multilingual LLMs

Idioms Understood, Yet Translated Literally: Diagnosing Literal Translation Bias in Multilingual LLMs

Yanming Sun, Runzhe Zhan, Shuo Wang, Jingkun Ma, Yu Chao, Lidia S. Chao, Derek F. Wong

arxiv

LLMs can recognize and interpret idioms yet still translate them literally, and we diagnose this literal translation bias through a cascaded framework and representation-level intervention.

Lean2Isabelle: Factorized Cross-Assistant Proof Translation

Lean2Isabelle: Factorized Cross-Assistant Proof Translation

Guanyu Liu*, Jingkun Ma*, Derek F. Wong

ICML 2026 Workshop@AI4Math

Lean2Isabelle factorizes Lean-to-Isabelle proof translation into statement prediction and proof generation, showing that verifier-guided RL and Lean proof inputs improve semantically accepted cross-assistant proof migration.

From Local Success to Composite Failure: Diagnosing Local-to-Composite Transfer in Mathematical Reasoning

From Local Success to Composite Failure: Diagnosing Local-to-Composite Transfer in Mathematical Reasoning

Jingkun Ma*, Haoran Wang*, Yang Li, Rongsheng Zhang, Derek F. Wong

arxiv

We introduce a paired diagnostic framework showing that mathematical reasoning models can solve local components yet fail when those components are bound into composed dependency structures.

News

2026-06

Two of our papers, FIPO-Prover and Lean2Isabelle, have been accepted to the ICML 2026 Workshop AI4Math, with FIPO-Prover selected as a Spotlight. We will present them on site. See you in Seoul, Korea South Korea!

2026-05

Our paper VisAidMath: Benchmarking Visual-Aided Mathematical Reasoning has been accepted to ACL 2026. We will also present it on site in San Diego, California, United States United States.

2026-06

Yang Li and Chao Tang are conducting Summer Research 2026 at Westlake University Westlake.

2026-06

Ziyan Hong and Yangshen Pan are conducting Summer Research 2026 at the University of Macau UM.

2026-05

Rongsheng Zhang has been admitted to the M.S. program at Johns Hopkins University JHU.

2026-05

Lingyi Ou has been admitted to the M.S. program at HKUST HKUST.

2026-05

Xingchen Chen has been admitted to the M.S. program at the University of Macau UM.

2026-05

Zhexuan Zhang has been admitted to the M.S. program at CUHK CUHK.

2025-12

VisAidMath received the Best Paper Award at the first VLM4RWD Workshop at NeurIPS 2025.

2025-07

Wenxi Ao and Lingyi Ou conducted Summer Research 2025 at the University of Macau UM.

2025-08

Started my Ph.D. in Computer Science at the University of Macau.

2023-05

Released Yu Sheng, a human-in-the-loop Classical Chinese poetry generation system, at EACL 2023 Demo.