Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
E-commerce Agent Interaction on WebShop
Loading...
70.4
Average Reward
MCTS
-2.468952
16.448949
35.36685
54.284751
Jul 25, 2025
Sep 15, 2025
Nov 6, 2025
Dec 28, 2025
Feb 18, 2026
Apr 11, 2026
Jun 2, 2026
Average Reward
Updated 1mo ago
Evaluation Results
Method
Method
Links
Average Reward
MCTS
Base LLM=Llama2-70B
2025.07
70.4
BPO
Approach=System-2, Eva...
2025.08
67.45
Tree DPO
Base LLM=Llama2-70B
2025.07
65.9
SFT
Base LLM=Llama2-70B
2025.07
65.1
MPO
Approach=System-2, Eva...
2025.08
55.2
Qwen-2.5-7B-Instruct
Approach=System-1, Eva...
2025.08
54.28
SFT
Approach=System-2, Eva...
2025.08
52.94
ETO
Approach=System-2, Eva...
2025.08
52.08
o3-mini
Approach=System-2, Eva...
2025.08
49.02
Qwen-3-Thinking
Approach=System-2, Eva...
2025.08
42.16
Deepseek-R1
Approach=System-2, Eva...
2025.08
40.31
Llama-3.1-8B-Instruct
Approach=System-1, Eva...
2025.08
33.93
DELTAMEM
LLM=DeepSeek-V4-flash,...
2026.06
0.6254
Synapse
LLM=DeepSeek-V4-flash,...
2026.06
0.589
No Memory
LLM=DeepSeek-V4-flash,...
2026.06
0.3674
RBank
LLM=DeepSeek-V4-flash,...
2026.06
0.3498
AWM
LLM=DeepSeek-V4-flash,...
2026.06
0.3337
Feedback
Search any
task
Search any
task