| A framework for benchmarking Class-out-of-distribution detection and its application to ImageNet I Galil, M Dabbah, R El-Yaniv International Conference on Learning Representations, 0 | 37* | |
| What Can we Learn From The Selective Prediction And Uncertainty Estimation Performance Of 523 Imagenet Classifiers? I Galil, M Dabbah, R El-Yaniv International Conference on Learning Representations, 0 | 26* | |
| Disrupting Deep Uncertainty Estimation Without Harming Accuracy I Galil, R El-Yaniv Advances in Neural Information Processing Systems 34, 21285-21296, 2021 | 16 | 2021 |
| Which models are innately best at uncertainty estimation? I Galil, M Dabbah, R El-Yaniv arXiv preprint arXiv:2206.02152, 2022 | 4 | 2022 |
| Hierarchical selective classification S Goren, I Galil, R El-Yaniv Advances in Neural Information Processing Systems 37, 111047-111073, 2025 | 1 | 2025 |
| Puzzle: Distillation-Based NAS for Inference-Optimized LLMs A Bercovich, T Ronen, T Abramovich, N Ailon, N Assaf, M Dabbah, I Galil, ... arXiv preprint arXiv:2411.19146, 2024 | 1 | 2024 |
| No Data, No Optimization: A Lightweight Method To Disrupt Neural Networks With Sign-Flips I Galil, M Kimhi, R El-Yaniv arXiv preprint arXiv:2502.07408, 2025 | | 2025 |
| Padding Tone: A Mechanistic Analysis of Padding Tokens in T2I Models M Toker, I Galil, H Orgad, R Gal, Y Tewel, G Chechik, Y Belinkov arXiv preprint arXiv:2501.06751, 2025 | | 2025 |
| Disrupting Deep Uncertainty Estimation Without Harming Accuracy Supplementary Material I Galil, R El-Yaniv | | |
Ran El-YanivProfessor of Computer Science, Technion - Israel Institute of Technology. Chief Scientist - Deci AIVerified email at cs.technion.ac.il