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..
Fast Computation and Optimization for Opinion-Based Quantities of Friedkin-Johnsen Model
Authors: Haoxin Sun, Yubo Sun, Xiaotian Zhou, Zhongzhi Zhang
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on real-world networks demonstrate that our algorithms are both accurate and efficient, outperforming state-of-the-art methods and scaling effectively to large-scale networks. |
| Researcher Affiliation | Academia | Haoxin Sun, Yubo Sun, Xiaotian Zhou, Zhongzhi Zhang College of Computer Science and Artificial Intelligence Fudan University EMAIL, EMAIL |
| Pseudocode | Yes | The procedure is detailed in Algorithm PFS (Partial Rooted Forest Sampling); due to space constraints, the pseudocode is provided in the appendix. [...] Algorithm 1: PF-QE (Partial Rooted Forest Method for QE ) provides the pseudocode [...] Algorithm 2: PF-OPMIN(Partial Rooted Forest Method for Op Min) provides the pseudocode [...] Algorithm 3: PF-PDMIN [...] Appendix A.1 Pseudocode of Algorithm PFS Algorithm 4: PFS(G, S, l) |
| Open Source Code | Yes | Our source code is publicly available on https://github.com/Haoxin Sun98/FJ-PF. [...] We provide open access to the code at the anonymous link: https://github. com/Haoxin Sun98/FJ-PF. This URL is included in the submission, and the data used in our experiments are all publicly available. |
| Open Datasets | Yes | The datasets for chosen real networks are accessed publicly through KONECT [29] and SNAP [34]. |
| Dataset Splits | No | The paper does not explicitly provide training/test/validation dataset splits. It discusses sampling parameters (e.g., p = 10,000 samples) and variations for evaluation purposes, rather than traditional data partitioning for training and testing models. |
| Hardware Specification | Yes | All experiments are carried out using the Julia programming language within a computational environment equipped with a 2.5 GHz Intel E5-2682v4 CPU and 256GB of primary memory. |
| Software Dependencies | No | The paper mentions using the "Julia programming language" but does not specify a version number or any other software dependencies with version numbers. |
| Experiment Setup | Yes | We set l = 1000, p = 10000, and k = 50 for our algorithms. [...] To ensure a fair comparison, we set the number of forests in [44] to 1000, consistent with our experimental setup, and execute both algorithms in parallel using 8 threads. [...] In our experiments, we follow the parameter setting in [59], setting the size of the candidate edge set EC to 10^4. And we use 20 iterations for the JL lemma. |