| Gemini: a family of highly capable multimodal models G Team, R Anil, S Borgeaud, JB Alayrac, J Yu, R Soricut, J Schalkwyk, ... arXiv preprint arXiv:2312.11805, 2023 | 6999 | 2023 |
| Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context G Team, P Georgiev, VI Lei, R Burnell, L Bai, A Gulati, G Tanzer, ... arXiv preprint arXiv:2403.05530, 2024 | 3448 | 2024 |
| Gemma 2: Improving open language models at a practical size G Team, M Riviere, S Pathak, PG Sessa, C Hardin, S Bhupatiraju, ... arXiv preprint arXiv:2408.00118, 2024 | 1816 | 2024 |
| Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities G Comanici, E Bieber, M Schaekermann, I Pasupat, N Sachdeva, I Dhillon, ... arXiv preprint arXiv:2507.06261, 2025 | 1217 | 2025 |
| BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning A Kirsch, J van Amersfoort, Y Gal NeurIPS 2019, 2019 | 878 | 2019 |
| Uncertainty estimation using a single deep deterministic neural network J van Amersfoort, L Smith, YW Teh, Y Gal International Conference on Machine Learning, 2020 | 810 | 2020 |
| Variational Recurrent Auto-Encoders O Fabius, J van Amersfoort ICLR 2015 Workshop, 2014 | 429 | 2014 |
| Deep deterministic uncertainty: A new simple baseline J Mukhoti, A Kirsch, J Van Amersfoort, PHS Torr, Y Gal Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern …, 2023 | 407* | 2023 |
| On feature collapse and deep kernel learning for single forward pass uncertainty J Van Amersfoort, L Smith, A Jesson, O Key, Y Gal arXiv preprint arXiv:2102.11409, 2021 | 209* | 2021 |
| Plex: Towards reliability using pretrained large model extensions D Tran, J Liu, MW Dusenberry, D Phan, M Collier, J Ren, K Han, Z Wang, ... arXiv preprint arXiv:2207.07411, 2022 | 150 | 2022 |
| Transformation-based models of video sequences J van Amersfoort, A Kannan, MA Ranzato, A Szlam, D Tran, S Chintala arXiv preprint arXiv:1701.08435, 2017 | 95 | 2017 |
| Frame interpolation with multi-scale deep loss functions and generative adversarial networks J Van Amersfoort, W Shi, A Acosta, F Massa, J Totz, Z Wang, J Caballero arXiv preprint arXiv:1711.06045, 2017 | 69* | 2017 |
| Prospect pruning: Finding trainable weights at initialization using meta-gradients M Alizadeh, SA Tailor, LM Zintgraf, J van Amersfoort, S Farquhar, ... International Conference on Learning Representations, 2022 | 64 | 2022 |
| Causal-bald: Deep bayesian active learning of outcomes to infer treatment-effects from observational data A Jesson, P Tigas, J van Amersfoort, A Kirsch, U Shalit, Y Gal Advances in Neural Information Processing Systems 34, 30465-30478, 2021 | 48 | 2021 |
| Mixtures of large-scale dynamic functional brain network modes C Gohil, E Roberts, R Timms, A Skates, C Higgins, A Quinn, U Pervaiz, ... NeuroImage 263, 119595, 2022 | 44 | 2022 |
| Deep deterministic uncertainty for semantic segmentation J Mukhoti, J Van Amersfoort, PHS Torr, Y Gal arXiv preprint arXiv:2111.00079, 2021 | 37 | 2021 |
| Single shot structured pruning before training J van Amersfoort, M Alizadeh, S Farquhar, N Lane, Y Gal arXiv preprint arXiv:2007.00389, 2020 | 26 | 2020 |
| Can convolutional ResNets approximately preserve input distances? A frequency analysis perspective L Smith, J van Amersfoort, H Huang, S Roberts, Y Gal arXiv preprint arXiv:2106.02469, 2021 | 11 | 2021 |
| Decomposing representations for deterministic uncertainty estimation H Huang, J van Amersfoort, Y Gal arXiv preprint arXiv:2112.00856, 2021 | 9 | 2021 |
| Deep hashing using entropy regularised product quantisation network J Schlemper, J Caballero, A Aitken, J van Amersfoort arXiv preprint arXiv:1902.03876, 2019 | 6 | 2019 |
Yarin GalProfessor of Machine Learning, University of OxfordVerified email at cs.ox.ac.uk
