Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Interactive Decision Making on SciWorld
Loading...
67.3
Score
AMC (ReAct)
-1.34
16.48
34.3
52.12
Jun 3, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
AMC (ReAct)
Policy Model=GPT-4.1-m...
2026.06
67.3
Best-of-15 (ReAct)
Policy Model=GPT-4.1-mini
2026.06
61.6
AMC (ReAct)
Policy Model=GPT-5.1,...
2026.06
59.7
Best-of-15 (ReAct)
Policy Model=GPT-5.1
2026.06
53.3
AMC (ReflAct)
Policy Model=Llama-3.1...
2026.06
37.6
AMC (ReAct)
Policy Model=Llama-3.1...
2026.06
34.7
Best-of-15 (ReflAct)
Policy Model=Llama-3.1-8B
2026.06
34.7
Best-of-15 (ReAct)
Policy Model=Llama-3.1-8B
2026.06
31.1
ReAct
Policy Model=GPT-4.1-mini
2026.06
25
ReAct
Policy Model=GPT-5.1
2026.06
9
ReflAct
Policy Model=Llama-3.1-8B
2026.06
5.1
ReAct
Policy Model=Llama-3.1-8B
2026.06
1.3
Feedback
Search any
task
Search any
task