Benchmark Leaderboards
Loading leaderboardsβ¦
Loading leaderboardsβ¦
Track which models hold the current SOTA on the benchmarks the field actually uses. Click a column header to sort; the leading row is highlighted.
Ranking quality (NDCG@20) on the Amazon-Book implicit-feedback benchmark, same canonical LightGCN split as the Recall@20 board. Transcribed from the SimGCL paper (Table 5). Higher is better.
Metric: NDCG@20 β higher is better Β· Source β
| # β² | Model / Paper | NDCG@20 | Ξ vs SOTA | Links |
|---|---|---|---|---|
| 1 | SimGCLSOTA | 0.041 | β | github.com β |
| 2 | SGL | 0.0379 | -0.00 | github.com β |
| 3 | MixGCF | 0.0378 | -0.00 | github.com β |
| 4 | DNN+SSL | 0.0337 | -0.01 | github.com β |
| 5 | LightGCN | 0.0315 | -0.01 | github.com β |
| 6 | Mult-VAE | 0.0315 | -0.01 | github.com β |