Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Embodied Task Planning on VirtualHome Novel Apartment (Unseen)
Loading...
82.9
Simple Success Rate
GPT3.5-MCTS
-3.316
19.067
41.45
63.833
May 23, 2023
Simple Success Rate
Comprehensive Success Rate
Novel Simple Success Rate
Novel Comprehensive Success Rate (Config 2)
Novel Comprehensive Success Rate (Config 3)
Updated 4d ago
Evaluation Results
Method
Method
Links
Simple Success Rate
Comprehensive Success Rate
Novel Simple Success Rate
Novel Comprehensive Success Rate (Config 2)
Novel Comprehensive Success Rate (Config 3)
GPT3.5-MCTS
backbone=GPT-3.5, sear...
2023.05
82.9
71.9
79.3
70.4
38.8
GPT3.5 Policy
backbone=GPT-3.5, mode...
2023.05
74.3
43.3
67.8
54
6.9
finetuned GPT2 policy
backbone=GPT-2, traini...
2023.05
65.5
39.9
33.4
12.8
1.1
UCT
search=UCT, commonsens...
2023.05
0
0
0
0
0
Feedback
Search any
task
Search any
task