| PRIMA: general and precise neural network certification via scalable convex hull approximations MN Müller, G Makarchuk, G Singh, M Püschel, M Vechev Proceedings of the ACM on Programming Languages 6 (POPL), 1-33, 2022 | 148* | 2022 |
| Complete verification via multi-neuron relaxation guided branch-and-bound C Ferrari, MN Muller, N Jovanovic, M Vechev The Tenth International Conference on Learning Representations, 2022 (ICLR'22), 2022 | 127 | 2022 |
| First three years of the international verification of neural networks competition (VNN-COMP) C Brix, MN Müller, S Bak, TT Johnson, C Liu International Journal on Software Tools for Technology Transfer 25 (3), 329-339, 2023 | 95 | 2023 |
| The third international verification of neural networks competition (vnn-comp 2022): summary and results MN Müller, C Brix, S Bak, C Liu, TT Johnson arXiv preprint arXiv:2212.10376, 2022 | 75 | 2022 |
| Certified training: Small boxes are all you need MN Müller, F Eckert, M Fischer, M Vechev The Eleventh International Conference on Learning Representations (ICLR'23), 2022 | 62 | 2022 |
| Boosting randomized smoothing with variance reduced classifiers MZ Horváth, MN Müller, M Fischer, M Vechev The Tenth International Conference on Learning Representations, 2022 (ICLR'22), 2021 | 52 | 2021 |
| Taps: Connecting certified and adversarial training Y Mao, MN Müller, M Fischer, M Vechev Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS'23), 2023 | 26* | 2023 |
| SWT-bench: Testing and validating real-world bug-fixes with code agents N Mündler, M Müller, J He, M Vechev Advances in Neural Information Processing Systems 37, 81857-81887, 2024 | 18* | 2024 |
| Robust and Accurate--Compositional Architectures for Randomized Smoothing MZ Horváth, MN Müller, M Fischer, M Vechev arXiv preprint arXiv:2204.00487, 2022 | 17 | 2022 |
| Certify or predict: Boosting certified robustness with compositional architectures MN Müller, M Balunović, M Vechev The Ninth International Conference on Learning Representations, 2021 (ICLR'21), 2021 | 15 | 2021 |
| Evading data contamination detection for language models is (too) easy J Dekoninck, MN Müller, M Baader, M Fischer, M Vechev arXiv preprint arXiv:2402.02823, 2024 | 14 | 2024 |
| Abstract interpretation of fixpoint iterators with applications to neural networks MN Müller, M Fischer, R Staab, M Vechev Proceedings of the ACM on Programming Languages 7 (PLDI), 786-810, 2023 | 14* | 2023 |
| Understanding certified training with interval bound propagation Y Mao, MN Müller, M Fischer, M Vechev arXiv preprint arXiv:2306.10426, 2023 | 13 | 2023 |
| Certified robustness to data poisoning in gradient-based training P Sosnin, MN Müller, M Baader, C Tsay, M Wicker arXiv preprint arXiv:2406.05670, 2024 | 9 | 2024 |
| Mitigating catastrophic forgetting in language transfer via model merging A Alexandrov, V Raychev, MN Müller, C Zhang, M Vechev, K Toutanova arXiv preprint arXiv:2407.08699, 2024 | 8 | 2024 |
| Spear: Exact gradient inversion of batches in federated learning DI Dimitrov, M Baader, M Müller, M Vechev Advances in Neural Information Processing Systems 37, 106768-106799, 2024 | 6 | 2024 |
| Expressivity of ReLU-Networks under Convex Relaxations M Baader, MN Müller, Y Mao, M Vechev arXiv preprint arXiv:2311.04015, 2023 | 6 | 2023 |
| The third international verification of neural networks competition (VNN-COMP 2022): summary and results (2022) MN Müller, C Brix, S Bak, C Liu, TT Johnson URL https://arxiv. org/abs/2212.10376, 2022 | 6 | 2022 |
| Constat: Performance-based contamination detection in large language models J Dekoninck, M Müller, M Vechev Advances in Neural Information Processing Systems 37, 92420-92464, 2024 | 5 | 2024 |
| Efficient Certified Training and Robustness Verification of Neural ODEs M Zeqiri, MN Müller, M Fischer, M Vechev The Eleventh International Conference on Learning Representations (ICLR'23), 2023 | 5* | 2023 |
Martin VechevFull Professor of Computer Science, ETH Zurich; Scientific Director, INSAIT;Verified email at inf.ethz.ch
