Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
Online and Consistent Correlation Clustering
Authors: Vincent Cohen-Addad, Silvio Lattanzi, Andreas Maggiori, Nikos Parotsidis
ICML 2022 | Venue PDF | LLM Run Details | Input Tokens: 30,134 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,340 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally we show experimentally that our algorithm achieves better performances than stateof-the-art algorithms on real world data. 5. Experiments |
| Researcher Affiliation | Collaboration | 1Google 2EPFL, Lausanne, Switzerland. |
| Pseudocode | Yes | Algorithm 1 AGREEMENT Algorithm 2 ONLINE AGREEMENT (Agree-On) |
| Open Source Code | Yes | Our code is written in C++ and is available online3. https://github.com/google-research/google-research/tree/master/online_correlation_clustering |
| Open Datasets | Yes | Our datasets are obtained from SNAP (Leskovec & Krevl, 2014); their basic characteristics are summarized in Table 1. |
| Dataset Splits | No | The paper describes an online setting where nodes arrive sequentially and does not specify traditional train/validation/test dataset splits with percentages or sample counts. |
| Hardware Specification | Yes | We run our experiments on a e2-standard-16 Google Cloud instance, with 16 cores, 2.20GHz Intel(R) Xeon(R) processor, and 64 Gi B main memory. |
| Software Dependencies | No | The paper states, 'Our code is written in C++,' but does not provide specific version numbers for any libraries, frameworks, or compilers used. |
| Experiment Setup | Yes | In our experiments we set the parameters β = λ = 0.2, as this setting exhibited the best behavior in (Cohen-Addad et al., 2021). That is, whenever a new node arrives it is inserted in the preexisting random order at a random position (the relative order of the previously arrived nodes remains unchanged). |