Scholar
Da Li
Google Scholar ID: RPvaE3oAAAAJ
Samsung AI Centre Cambridge
Citations & Impact
Citations
5,824
H-index
20
i10-index
25
Publications
20
Co-authors
38
list available
Publications
Resume (English only)
Academic Achievements
- TFAR: A Training-Free Framework for Autonomous Reliable Reasoning in Visual Question Answering (Accepted by TMLR)
- ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning (ICLR 2025 Poster)
- On the Limitations of General Purpose Domain Generalisation Methods (Submitted to ICLR 2025)
- Recurrent Early Exits for Federated Learning with Heterogeneous Clients (ICML 2024 Poster)
- Neural Fine-Tuning Search for Few-Shot Learning (ICLR 2024 oral)
- Vision-Language Subspace Prompting (Submitted to ICLR 2024)
- Worst-case Feature Risk Minimization for Data-Efficient Learning (Accepted by TMLR)
- FedL2P: Federated Learning to Personalize (NeurIPS 2023 poster)
- Better Practices for Domain Adaptation (AutoML 2023 MainTrack)
- Label Calibration for Semantic Segmentation Under Domain Shift (ICLR 2023 Workshop on Trustworthy ML Poster)
Research Experience
- Visiting Lecturer, EECS, Queen Mary, University of London, 2025 - Present
- Sr. Research Scientist, Samsung AI Center, Cambridge, 2019 - Present
- Visiting Lecturer, University of Surrey, 2023 - 2024
- Visiting Scholar, University of Edinburgh, 2020 - 2024
- Research Assistant, Queen Mary, University of London, 2016 - 2019
- Software Engineer, Autodesk, 2014 - 2016
Education
- PhD from Queen Mary, University of London, 2016-2019, under the supervision of Yi-Zhe Song, Timothy Hospedales, and Tao Xiang.
Background
- Research interests include RL, Agentic AI, GenAI, Multimodal/VLM, Federated Learning, Few-shot Learning, Domain Adaptation, Machine Learning, Domain Generalization, and Computer Vision.
Miscellany
- Personal Links: Homepage, Google Scholar, DBLP, ORCID
