Webshop
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
WebShop Drift II v1.0
95Success Rate
9
WebShop Drift I v1.0
90Success Rate
9
WebShop Source v1.0
100Success Rate
9
WebShop semantic shift Implicit Drift (Source)
50Score
9
WebShop (In-domain)
100Success Rate
8
Webshop
73.4P@1
8
WebShop (test)
0.8Average Reward
8
WebShop (50 instructions)
82.1Score
8
WebShop
77.88Avg@8 Success Rate
7
WebShop
0.185ECE (End-State)
7
WebShop (test)
46Score
6
WebShop (iter1)
0.9rho
6
WebShop hundred-product catalog setup
90.2Mean Reward
6
WebShop
87E2E Success Rate
6
WebShop (Seen)
70Success Rate
6
WebShop (test)
81.4Reward
6
WebShop
99.3Utility Score
5
Webshop
82.1Score
5
WebShop OOD
79Score
4
WebShop (OOD)
49.6Score
4
WebShop
77.25Average Reward
4
WebShop Environment Injection (test)
87E2E Success
4
WebShop (post-training evaluation)
14.8Rollout Length
3
WebShop 1000 (test)
81.2Success Rate
3
WebShop 40k (test)
0.741Average Reward
3