MorphoSuffPreprint · 2026
When label-free morphology is sufficient for targeted cellular measurements
Tests when morphology-derived measurements preserve cellular responses and the scientific decisions based on them.
AI FOR SCIENCE · SCIENTIFIC AGENTS · GRAPH LEARNING
Mengran Li 李孟燃
Ph.D. Student in Control Science & Engineering
Sun Yat-sen University ↗
My research spans AI for Science, graph and hypergraph learning, temporal modeling, and bioinformatics. I now focus on cellular predictions: when they support scientific decisions, how models improve with feedback, and whether they use the intended inputs.
I am advised by Prof. Ronghui Zhang and Prof. Yong Zhang. Previously, I interned at Baidu, SIAT, Westlake University, and CAIR.
Open to opportunities
Research collaborations, internships, and AI for Science roles.
limengran1998@163.com ↗01 / UPDATES
News & milestones
- Awarded the National Scholarship for Ph.D. Students, and the Presidential Scholarship of Sun Yat-sen University.
- One collaborative paper accepted by IEEE TNNLS.
- Two papers accepted by Renewable Energy, including one as corresponding author and one collaborative work.
- One paper accepted by IEEE/CAA Journal of Automatica Sinica.
- Recognized as a Gold Reviewer for ICML 2026.
View all 18 updates ＋
- One collaborative paper accepted by ICML 2026.
- One paper accepted by IEEE TCYB.
- One paper accepted by IEEE TNNLS.
- Presented our work as an Oral Presentation at AAAI 2026 in Singapore: Learning Cell-Aware Hierarchical Multi-Modal Representations for Robust Molecular Modeling.
- One paper accepted by Pattern Recognition (PR).
- Two collaborative papers were accepted by Nature Communications and Communications Biology, respectively.
- Research reported by media: Analysis of CHMR Framework.
- One paper accepted by AAAI 2025 (Oral).
- Awarded the National Scholarship for Ph.D. Students.
- Awarded the Presidential Scholarship of Sun Yat-sen University.
- Two papers (Redundancy Is Not What You Need... and SCAE...) selected as ESI Highly Cited Papers.
- One survey paper on Graph LLMs accepted by Expert Systems with Applications (ESWA).
- Awarded the SYSU Hong Kong/Macau Exchange Funding. Joint training at Westlake University and CAIR, HK (Supervisor: Prof. Stan Z. Li, Prof. Zhen Lei and Prof. Jiebo Luo).
02 / SELECTED RESEARCH
Selected Publications 代表作
Representative work in cellular AI and graph learning. Click any figure to enlarge it.
CellScientistPreprint · 2026
CellScientist: From Execution Feedback to Auditable Model-Revision Trajectories for Cellular Perturbation Prediction
Uses execution feedback to revise cellular-response models under a fixed evaluation protocol, recording each design change and outcome.
CellAuditPreprint · 2026
Discover, Falsify, Revise: Auditing Input-Use Claims from Source Code to Predictive Contribution in Agent-Discovered Cell Models
Checks whether cellular-response models actually use perturbation inputs and whether those inputs improve prediction, then uses failures to guide revision.
AttriReBoostIEEE Transactions on Cybernetics (IEEE TCYB 2026)
AttriReBoost: A Gradient-Free Propagation Optimization Method for Cold-Start Mitigation in Attribute Missing Graphs
We propose ARB, a gradient-free propagation framework for attribute-missing graphs that mitigates cold-start via boundary redefinition and virtual edges, achieving accurate reconstruction with strong scalability.
APPIEEE TNNLS 2026
Adaptive Prototype-Guided Personalized Propagation for Heterophilic Graphs With Missing Data
Combines semantic prototypes, virtual edges, and adaptive fusion to learn from heterophilic graphs with missing features.
CHMRAAAI 2026 (Oral)
Learning Cell-Aware Hierarchical Multi-Modal Representations for Robust Molecular Modeling
Models molecular structure and cellular responses together, using hierarchical representations to improve molecular property prediction.
03 / PUBLICATION RECORD
Full Publication List
31 publications
23 journals5 conferences3 preprints
SFAR: Semantic Fusion Attribute Recovery for Text Attribute Missing Graphs via Large Language Model Knowledge Generalization
04 / RECOGNITION
Honors & awards
- National Scholarship (2026, 2025, 2022)
- Presidential Scholarship, SYSU (2026, 2025)
- Outstanding Master's Thesis, BJUT (2023)
- Xiaomi Special Award (Top 10), BJUT (2022)
05 / ACADEMIC COMMUNITY
Service
Reviewer: IJCV, IEEE TPAMI, ICML, NeurIPS, ICLR, AAAI, ACM MM; IEEE TNNLS, IEEE TKDE, ACM TKDD, TOIS; Pattern Recognition, Information Fusion, etc.
GET IN TOUCH
Let’s work on scientific AI.
I welcome conversations about research collaborations and opportunities.