| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Addition | 10-digit addition 3 terms | Token Accuracy89.56 | 14 | |
| Addition | Addition | Accuracy99.5 | 7 | |
| Addition | Addition synthetic (val) | Skill 1 Score0.016 | 7 | |
| Token-level Addition | Addition T3 (val) | Token Accuracy89.56 | 6 | |
| Binary Addition | Addition (OOD lengths 20-60) | OOD Performance45 | 4 | |
| Arithmetic Addition | ADDITIONBase-5 (test) | Accuracy23.5 | 4 | |
| Arithmetic Addition | ADDITIONBase-3 (test) | Accuracy36.3 | 4 | |
| Arithmetic Addition | ADDITIONBase-2 (test) | Accuracy95.4 | 4 | |
| Addition | Addition 200 prompts Qwen3-8B (test) | Accuracy100 | 3 |