Amazon
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Amazon Toys and Games
5.21NDCG@5
24
Amazon Electronics
14.2NDCG@10
23
Amazon Books
0.126NDCG@10
23
Amazon
34.9ASR
22
Amazon Sports
0.204Recall@10
22
Amazon Video-Games
0.532NDCG@10
22
Amazon (test)
89.5F1-macro
22
Amazon (test)
36.57ROUGE-1 Score
22
Amazon Video
83.5HR@10
21
Amazon-Clothing
0.111Recall@10
21
Amazon Sports
12.26Recall@10
21
Amazon Sports and Outdoors
2.8Recall@5
21
Amazon Elec -> Cloth OOD #2
65.84HR@10
20
Amazon StandardCL (test)
60.22Accuracy
20
Amazon
100AUC
20
Amazon (test)
86Accuracy
20
Amazon zero-shot (test)
82.67AUROC
19
Amazon v1 (full)
0.651AUC
19
Amazon Beauty (test)
6.28BLEU-4
19
Amazon Beauty (test)
39.6NDCG@10
19
Amazon Electronics
14.3Recall
18
Amazon Games
6.09NDCG@10
18
Amazon Beauty
20.36NDCG@10
18
Amazon Pantry (test)
0.062NDCG@10
18
Amazon Electronic (IID)
8.93Recall@20
18