Position: Certified Robustness Does Not (Yet) Imply Model Security
Authors: Andrew Craig Cullen, Paul Montague, Sarah Monazam Erfani, Benjamin I. P. Rubinstein
ICML 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | This position paper is a call to arms for the certification research community, proposing concrete steps to address these fundamental challenges and advance the field toward practical applicability. |
| Researcher Affiliation | Collaboration | 1School of Computing and Information Systems, University of Melbourne, Australia 2DST Group, Adelaide, Australia. |
| Pseudocode | No | The paper describes theoretical concepts and arguments but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper is a position paper and does not present new methodology for which code would be released. It mentions 'Open DP' as an example of open development, but this is not code for the paper's own work. |
| Open Datasets | No | The paper references well-known datasets like MNIST, CIFAR-10, and ImageNet as 'key reference datasets' for validating research, but it does not conduct its own experiments using these datasets or provide access information for a dataset it specifically uses or creates. |
| Dataset Splits | No | The paper does not conduct experiments or analyze data, therefore, no dataset split information is provided. |
| Hardware Specification | No | The paper is a position paper discussing theoretical and conceptual challenges; it does not describe any experimental setup or specific hardware used for computations. |
| Software Dependencies | No | The paper is a position paper and does not describe any specific software dependencies with version numbers for its own work. It mentions 'Open DP' as a development example but not as a dependency for its own methodology. |
| Experiment Setup | No | The paper is a position paper and does not present any experimental results or methodology that would require details on experimental setup, hyperparameters, or training configurations. |