F1 Score on Long Context Reasoning on LongBench
68.5F1 ScoreAoT
Evaluation Results
| Method | Links | |
|---|---|---|
| AoT2026.04 | 68.5 | |
| WORC+AC2026.04 | 68.4 | |
| AgentChain2026.04 | 64.3 | |
| AFlow2026.04 | 61 | |
| FoTn (sampling size)=82026.04 | 59.1 | |
| CoT-SCn (sampling size)=52026.04 | 58.6 | |
| Self-Refine2026.04 | 58.2 | |
| CoT2026.04 | 57.6 | |
| Analogical Prompting2026.04 | 52.9 |