Multi-hop Question Answering on HotpotQA (Online Collection Split)
61.4Pass@1EXG-Reflexion
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| EXG-ReflexionBackbone=Qwen-Plus2026.05 | 61.4 | 67 | 1.79 | |
| ExpeLBackbone=Qwen-Plus2026.05 | 60 | 66.1 | 1.8 |
Loading the SOTA2 catalog…
| Method | Links | |||
|---|---|---|---|---|
| EXG-ReflexionBackbone=Qwen-Plus2026.05 | 61.4 | 67 | 1.79 | |
| ExpeLBackbone=Qwen-Plus2026.05 | 60 | 66.1 | 1.8 |