analiza aiqbench
This paper challenges the necessity of complex learned node embeddings in temporal link prediction by demonstrating that smoothed statistical counts can outperform state-of-the-art neural methods on most datasets. By reducing the model to just 9–13 parameters, it establishes a lightweight, interpretable baseline for evaluating future AI-driven graph forecasting systems. The results suggest that simple probabilistic modeling may be sufficient for many dynamic network tasks, offering a pragmatic alternative to heavy embedding-based architectures.
Do Temporal Link Predictors Need Learned Memory? A Smoothed-Count Baseline with a Handful of Parameters
arXiv:2609.35364v1 Announce Type: new Abstract: Many temporal link predictors summarize past interactions through learned node representations. We examine whether simple counts of recurring interaction patterns can provide competitive predictions without learning these representations. We propose a temporal link predictor based on statistical language modelling. It pools transition and co-occurrence counts across sources to predict links that a source has never formed. We smooth sparse estimates using destination frequencies or Kneser-Ney continuation counts. A shared log-linear rule combines these estimates with popularity, source history, and recency, without node embeddings. In our main evaluation, the model achieves the highest MRR among the compared methods on 7 out of 16 datasets from TGB and TGB-Seq. It also outperforms EdgeBank and Base3 on all 16 datasets and the heuristic family on 14. These gains extend to datasets designed to limit repeated edges. With only 9--13 learned parameters, our model provides a simple and competitive baseline for evaluating future neural temporal link predictors.
Title:Do Temporal Link Predictors Need Learned Memory? A Smoothed-Count Baseline with a Handful of Parameters
Abstract:Many temporal link predictors summarize past interactions through learned node representations. We examine whether simple counts of recurring interaction patterns can provide competitive predictions without learning these representations. We propose a temporal link predictor based on statistical language modelling. It pools transition and co-occurrence counts across sources to predict links that a source has never formed. We smooth sparse estimates using destination frequencies or Kneser-Ney continuation counts. A shared log-linear rule combines these estimates with popularity, source history, and recency, without node embeddings. In our main evaluation, the model achieves the highest MRR among the compared methods on 7 out of 16 datasets from TGB and TGB-Seq. It also outperforms EdgeBank and Base3 on all 16 datasets and the heuristic family on 14. These gains extend to datasets designed to limit repeated edges. With only 9--13 learned parameters, our model provides a simple and competitive baseline for evaluating future neural temporal link predictors.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2609.35364 [cs.LG]
(or arXiv:2609.35364v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.35364 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 265 cuvinte
