Mateja Jamnik
Professor of Artificial Intelligence
Develops human-like AI that combines automated reasoning with machine learning to make AI systems explainable, applied to personalised medicine and tutoring systems.
Mateja Jamnik's lab at Cambridge works on trustworthy & safe ML, reasoning & neuro-symbolic AI and reasoning & agents. 46 papers since 2023, 16 at top venues, mostly ICML and ICLR. 11 PhD students. 15 alumni. Applications close .
At a glance
- papers, 2023–2026
- 12/ year
- at T1 venues
- 35%
- PhD students
- 11
- MS students
- 0
- alumni listed
- 15
Papers by yearfrom the lab homepage, 2023–2026
- 10 papers2026 · T1 3 · collecting
- 8 papers2025 · T1 6
- 16 papers2024 · T1 5
- 12 papers2023 · T1 2
T1 venuesOtherVenues: ICML, ICLR, TMLR, AAAI, PNAS
Research focuspaper titles that use each topic's words
- Reasoning & neuro-symbolic AI2 papers
- Reasoning & agents2 papers
- Trustworthy & safe ML0 papers
- make0 papers
- tutoring systems0 papers
Paper list
From the lab homepage.
202610 papers
- Proceedings of the LAK26: International Learning Analytics and Knowledge Conference, LAKCan Micro-Behaviours be used for Graph Comprehension Assessment?
- ICML (T1)CB-SLICE: concept-based interpretable error slice discovery
- Advances in Neural Information Processing Systems 40: Annual Conference on Neural Information Processing Systems, NeurIPSHaoyan SKOP
- ICLR (T1)Hierarchical Concept-based Interpretable Models
- —Leiden declaration on artificial intelligence and mathematics
- Notices of the American Mathematical Society, 73(6):480-483Shaping the Future of Mathematics in the Age of AI
- Advances in Neural Information Processing Systems 40: Annual Conference on Neural Information Processing Systems, NeurIPSShubra interestingness
- ICLR (T1)TabStruct: Measuring Structural Fidelity of Tabular Data
- Communications of the ACM BlogThe Leiden Declaration: Mathematics, AI, and Making Our Values Explicit
- Advances in Neural Information Processing Systems 40: Annual Conference on Neural Information Processing Systems, NeurIPSXiangjian TabForge
20258 papers
- ICML (T1)Avoiding leakage poisoning: Concept interventions under distribution shifts
- TMLR (T1)Do concept bottleneck models respect localities?
- Computer Law and Security Review, 56:106111. ElsevierFuture themes in regulating artificial intelligence in investment management
- AAAI (T1)Measuring cross-modal interactions in multimodal models
- ICLR (T1)Multimodal lego: Model merging and fine-tuning across topologies and modalities in biomedicine
- AAAI (T1)Neural reasoning for sure through constructing explainable models
- ICML (T1)NMA-tune: Generating highly designable and dynamics aware protein backbones
- CITRExRO-FIGS: Efficient and expressive tree-based ensembles for tabular data
202416 papers
- CHI : CHI Conference on Human Factors in Computing Systems, 194:1--194:14. ACMA human information processing theory of the interpretation of visualizations: Demonstrating its utility
- Annual Conference of the Cognitive Science Society (CogSci ). Cognitive Science SocietyDecoding expertise: Exploring cognitive micro-behavioural measurements for visualization competence
- ICLR (T1)Dynamics-Informed Protein Design with Structure Conditioning
- Leonardis, A. et al (eds.), European Conference in Computer Vision, ECCV , Lecture Notes in Computer Science. SpringerEfficient bias mitigation without privileged information
- Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems, NeurIPSEnd-to-end ontology learning with large language models
- PNAS (T1)Evaluating language models for mathematics through interactions
- TMLR (T1)GCondNet: A novel method for improving neural networks on small high-dimensional tabular data
- Wooldridge, M. et al, (eds.), Thirty-Eighth AAAI Conference on Artificial Intelligence, AAAI , Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence, IAAI , Fourteenth Symposium on Educational Advances in Artificial Intelligence, EAAIGeneration of visual representations for multi-modal mathematical knowledge
- Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems, NeurIPSHEALNet: Multimodal fusion for heterogeneous biomedical data
- Annual Conference of the Cognitive Science Society (CogSci ). Cognitive Science SocietyIndex systems: Enumerating their forms and explaining their diversity with representational interpretive structure theory
- Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems, NeurIPSMulti-language diversity benefits autoformalization
- CICMOruga: Implementation and use of representational systems theory
202312 papers
- Colarusso, F., Cheng, P.C.H., Garcia Garcia, G., Stockdill, A., Raggi, D. and Jamnik, MA novel interaction for competence assessment using micro-behaviors: Extending CACHET to graphs and charts
- TML4HCGXplain: Rule-based deep neural network explanations using dual linear programs
- Longo, L., (ed.), Explainable Artificial Intelligence, xAIConcept Distillation in Graph Neural Networks
- Lishkova, Y., Scherer, P., Ridderbusch, S., Jamnik, M., Lio, P., Ober-Bloebaum, S. and Offen, CDiscrete lagrangian neural networks with automatic symmetry discovery
- International Conference on Learning Representations (ICLR). OpenReview.netDraft, sketch, and prove: Guiding formal theorem provers with informal proofs
- AIES '23: Proceedings of the AAAI/ACM Conference on AI, Ethics, and SocietyHuman Uncertainty in Concept-Based AI Systems
- IEEE Symposium on Visual Languages and Human-Centric Computing, VL/HCCHuman visual consistency-checking in the real world ontologies
- ICML (T1)Interpretable Neural-Symbolic Concept Reasoning
- Oh, Alice at al (eds.), Advances in Neural Information Processing Systems 37: Annual Conference on Neural Information Processing Systems , NeurIPSLearning to receive help: Intervention-aware concept embedding models
- TMLR (T1)TabCBM: Concept-based interpretable neural networks for tabular data
- Espinosa Zarlenga, M., Shams, Z., Barbiero, P., Kazhdan, D., Bhatt, U., Weller, A. and Jamnik, MTowards robust metrics for concept representation evaluation
- Margeloiu, A., Simidjievski, N., Lio, P. and Jamnik, MWeight predictor network with feature selection for small sample tabular biomedical data
Where alumni went15 listed, 0% to industry
Industry 0Academia 1
Also to
- University of Edinburgh
Apply to CambridgeMichaelmas 2027
PhD in Computer Science
- Round
- Funding round (8 Dec 2026)
- Opens
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- USMIT
Manish Raghavan
- Trustworthy & safe ML
- USUIUC
Halil Kilicoglu
- Knowledge, QA & retrieval
- USUCSD
Zhiting Hu
- Generative & foundation models
- USMichigan
Anhong Guo
- Human-AI collaboration & trust
- USStanford
Emma Brunskill
- Reinforcement learning
- GBEdinburgh
Elizabeth Polgreen
- Compilers & programming languages
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Cambridgecloses