kaft-neurips.git/figures/gen_fig1_diagnostic.py, branch main Unnamed repository; edit this file 'description' to name the repository. Initial release: GRAFT (KAFT) — NeurIPS 2026 submission code 2026-05-05T04:05:16+00:00 YurenHao0426 blackhao0426@gmail.com 2026-05-05T04:05:16+00:00 bd9333eda60a9029a198acaeacb1eca4312bd1e8 Topology-factorized Jacobian-aligned feedback for deep GNNs. Includes: - src/: GraphGrAPETrainer (KAFT) + BP / DFA / DFA-GNN / VanillaGrAPE baselines + multi-probe alignment estimator + dataset / sparse-mm utilities. - experiments/: 19 runners reproducing every figure / table in the paper. - figures/: 4 generators + the 4 PDFs cited in the report. - paper/: NeurIPS .tex and consolidated experiments_master notes. Smoke test: 50-epoch Cora GCN L=4 gives BP 77.3% / KAFT 79.0%. 
Topology-factorized Jacobian-aligned feedback for deep GNNs. Includes:
- src/: GraphGrAPETrainer (KAFT) + BP / DFA / DFA-GNN / VanillaGrAPE baselines
        + multi-probe alignment estimator + dataset / sparse-mm utilities.
- experiments/: 19 runners reproducing every figure / table in the paper.
- figures/: 4 generators + the 4 PDFs cited in the report.
- paper/: NeurIPS .tex and consolidated experiments_master notes.

Smoke test: 50-epoch Cora GCN L=4 gives BP 77.3% / KAFT 79.0%.