Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Integrated Defense for Resilient Graph Matching
Authors: Jiaxiang Ren, Zijie Zhang, Jiayin Jin, Xin Zhao, Sixing Wu, Yang Zhou, Yelong Shen, Tianshi Che, Ruoming Jin, Dejing Dou
ICML 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the robustness of our IDRGM model on real datasets against state-of-the-art algorithms.Empirical evaluation over real graph datasets demonstrates that the remarkable robustness of IDRGM against state-of-the-art graph matching methods and representative resilient Lipschitz-bound neural architectures. We report Hits@K (Yasar & C ataly urek, 2018; Fey et al., 2020) to evaluate and compare our model to previous lines of work, where Hits@K measures the proportion of correctly matched nodes ranked in the top-K list. |
| Researcher Affiliation | Collaboration | 1Auburn University, USA 2Peking University, China 3Microsoft Dynamics 365 AI, USA 4Kent State University, USA 5University of Oregon, USA 6Baidu Research, China. |
| Pseudocode | Yes | Algorithm 1 Expressive Parameter Estimation |
| Open Source Code | No | The paper states: "To our best knowledge, there are no other open-source defense baselines on graph matching available." However, it does not explicitly provide its own source code or a link to it for the methodology described. |
| Open Datasets | Yes | We will show the robustness of our IDRGM model for resilient graph matching over three datasets: autonomous systems (AS) (AS), CAIDA relationships datasets (CAI), and DBLP coauthor graphs (DBL), as shown in Table 1. Links provided in references: https://snap.stanford.edu/data/Oregon-2.html. https://snap.stanford.edu/data/as-Caida.html. http://dblp.uni-trier.de/xml/. |
| Dataset Splits | No | The dataset is divided into two disjoint sets: training data D and test data D .We randomly sample 30% of known matched node pairs as training data and the rest as test data. The paper specifies training and test data splits but does not explicitly mention a separate validation set. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU models, CPU types) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers. |
| Experiment Setup | No | In addition, more experiments, implementation details, and hyperparameter selection and setting are presented in Appendices A.2-A.4. These details are not explicitly present in the main text of the paper as required. |