Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embodied Household Tasks on ALFWorld Seen
Loading...
90.7
Average Accumulated Reward
Q-Evolve
-3.628
20.861
45.35
69.839
Jun 5, 2026
Average Accumulated Reward
Updated 1mo ago
Evaluation Results
Method
Method
Links
Average Accumulated Reward
Q-Evolve
Backbone=Llama-2-7B-Chat
2026.06
90.7
QLASS
Backbone=Llama-2-7B-Chat
2026.06
77.9
ETO
Backbone=Llama-2-7B-Chat
2026.06
68.6
RFT
Backbone=Llama-2-7B-Chat
2026.06
62.9
Best-of-N
Backbone=Llama-2-7B-Ch...
2026.06
62.1
SFT
Backbone=Llama-2-7B-Chat
2026.06
60
Reflexion
Prompting=ReAct
2026.06
45.7
GPT-4
Prompting=ReAct
2026.06
42.9
PPO
Backbone=Llama-2-7B-Chat
2026.06
22.1
GPT-3.5-Turbo
Prompting=ReAct
2026.06
7.9
Base Agent
Backbone=Llama-2-7B-Chat
2026.06
0
Feedback
Search any
task
Search any
task