Web Task Automation on WebArena-Lite
0.34Average RewardJEF-HINTER
Evaluation Results
| Method | Links | |
|---|---|---|
| JEF-HINTERBase Model=GPT-5-mini2025.10 | 0.34 | |
| DocumentationBase Model=GPT-5-mini2025.10 | 0.33 | |
| ReActBase Model=GPT-5-mini2025.10 | 0.32 | |
| JEF-HINTERBase Model=GPT-5-nano2025.10 | 0.27 | |
| ReActBase Model=GPT-5-nano2025.10 | 0.23 | |
| DocumentationBase Model=GPT-5-nano2025.10 | 0.2 |