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 [1].
Co-exposure Maximization in Online Social Networks
Authors: Sijing Tu, Cigdem Aslay, Aristides Gionis
NeurIPS 2020 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experimentally demonstrate the quality of our proposal on real-world social networks.Finally, we experimentally evaluate our algorithm on several real-world datasets and demonstrate its superiority over several baselines.We evaluate our method against different baselines on real-world networks. |
| Researcher Affiliation | Academia | Sijing Tu Department of Computer Science KTH Royal Institute of Technology Stockholm, Sweden EMAIL Cigdem Aslay Department of Computer Science Aarhus University Aarhus, Denmark EMAIL Aristides Gionis Department of Computer Science KTH Royal Institute of Technology Stockholm, Sweden EMAIL |
| Pseudocode | Yes | Algorithm 1: Pairs-Greedy |
| Open Source Code | Yes | Omitted proofs and implementation are provided as supplementary material. |
| Open Datasets | Yes | Datasets. We use the following networks: Flixster [6], Last.FM [5], Net HEPT [13], and Wiki Vote [30]. Basic statistics of these networks are reported in the supplementary. |
| Dataset Splits | No | The paper does not specify training, validation, and test dataset splits (e.g., percentages or exact counts) for reproducibility. It uses the term 'validation' in the context of a statistical test for estimating a lower bound, not for data partitioning. |
| Hardware Specification | Yes | Our experiments are performed on a server with a 2 × 10 core Xeon E5 2680 v2 2.80 GHz processor, with 256 GB memory. |
| Software Dependencies | No | The paper mentions various models and baselines but does not list specific software dependencies with version numbers (e.g., programming languages, libraries, or frameworks). |
| Experiment Setup | Yes | The confidence and accuracy parameters are set to ℓ = 1 and ϵ = 0.2. Our method takes as input seed budgets kr and kb... To ensure a fair comparison, we first execute Balance Exposure with varying k = 50, 100, 150, 200, and use the returned values kr and kb as input for the other methods. We fix τ = 2 and kr = 20. |