Agentic Reasoning on WebShop (test)
82.3Success Rate
RETROAGENT
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 82.3 | 88.9 | |
2026.03 | 78.9 | 87.6 | |
2026.03 | 72.8 | 84.4 | |
2026.03 | 72.7 | 85.2 | |
2026.03 | 69.3 | 81 | |
2026.03 | 66.9 | 75.5 | |
2026.03 | 65.7 | 80.3 | |
2026.03 | 61.7 | - | |
2026.03 | 46.9 | 67.8 | |
2026.03 | 37.5 | 58.1 | |
2026.03 | 28.8 | 58.1 | |
2026.03 | 19.5 | 46.2 | |
2026.03 | 17.6 | 42.5 | |
2026.03 | 9.2 | 29.5 | |
2026.03 | 0.8 | 4.5 |