Reproduce “Adapting to Evolving Graphs: A Scalable Framework for Dynamic Coarsening” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Abhishek Gupta, Manoj Kumar, Sarthak Kumar Singh, Ujjwal Yadav, Yifan Sun, Sandeep Kumar
The authors introduce dynamic graph coarsening that initializes node-to-supernode assignments once and updates them across discrete-time snapshots. DIGC keeps old assignments fixed; ACNR can remap them as topology changes. On ACM, DIGC plus GCN classification takes 0.19 seconds versus 1.04 seconds without coarsening. ACNR accuracy is 86.49%, close to recomputed FGC’s 86.74%, in citation graphs that mainly grow.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.