MSR: A Multifaceted Self-Retrieval Framework for Microscopic Cascade Prediction

Authors: Dongsheng Hong, Chao Chen, Xujia Li, Shuhui Wang, Wen Lin, Xiangwen Liao

AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirical evaluations on public datasets demonstrate that this framework significantly outperforms baselines on Hits@κ and MAP@κ, affirming its enhanced ability. ... Empirical Validation: Our method shows significant improvements over baseline models, with an average increase of 6.0% on Hits@κ and 4.4% on MAP@κ across three public datasets. Additionally, the ablation studies, visualizations, and parameter analysis further validate the effectiveness and robustness of each component.
Researcher Affiliation Academia 1 College of Computer and Data Science, Fuzhou University, China 2 School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), China 3 Hong Kong University of Science and Technology, Hong Kong SAR, China 4 College of Computer and Control Engineering, Minjiang University, China EMAIL, EMAIL, EMAIL, EMAIL
Pseudocode No The paper describes its methodology using mathematical formulations and descriptive text, but does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any explicit statements about releasing source code, nor does it provide links to a code repository.
Open Datasets Yes We experimented with the following datasets. Twitter (Hodas and Lerman 2014): Includes retweet records from 2010, with cascades formed by retweets. Memetracker (Leskovec, Backstrom, and Kleinberg 2009): Forms cascades of identical memes from social media, without explicit follower relationships. Douban (Zhong et al. 2012): Comprises cascades based on comments for movie and book reviews.
Dataset Splits Yes In alignment with previous studies (Sun et al. 2022; Wang et al. 2022; Li et al. 2023), we partitioned datasets into training, validation and test at a ratio of 8:1:1.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments, such as GPU or CPU models.
Software Dependencies No The paper mentions various components and techniques like GNNs, GRAU, GRU, GCN, etc., but does not specify software names with their version numbers (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes Given the importance of the temperature coefficient τ, we also conducted a parameter selection experiment to evaluate its impact on the model s performance. ... The result indicates that the model s performance stabilizes when τ is within the range of 0.05 to 0.1, with the optimal performance observed at τ = 0.02. ... Consequently, 0.05 is identified as yielding the best performance and is thus selected as the default setting for our method. ... The log-cross entropy loss is used for training.