Shopping Agent Task on Amazon Easy
87.5Kimi K2 Success RateSTRUCTUREDAGENT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| STRUCTUREDAGENTBackbone=Claude 3.5 Sonnet2026.03 | 87.5 | 75 | 87.5 | 83.3 | |
| StructuredAgentMemBackbone=Claude 3.5 Sonnet2026.03 | 80 | 72.5 | 82.5 | 78.3 | |
| BasicClaudeActionBackbone=Claude 3.5 Sonnet2026.03 | 77.5 | 60 | 77.5 | 71.7 | |
| AgentOccamBackbone=Claude 3.5 Sonnet2026.03 | 75 | 62.5 | 70 | 69.2 | |
| StePBackbone=Claude 3.5 Sonnet2026.03 | 62.5 | 57.5 | 62.5 | 60.8 |