Mathematical Reasoning on GSM8K (test) (Accuracy, Latency)
68.41GSM8K Accuracy (%)Explicit CoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Explicit CoTBackbone=LLAMA-3.2-3B-INSTRUCT2026.06 | 68.41 | 963.6 | |
| LOTUSBackbone=LLAMA-3.2-3B-INSTRUCT2026.06 | 68.13 | 140.8 |
Loading the SOTA2 catalog…