Code Generation on LCB monolingual
96.44RPRTLPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TLPON=16, Mode=Fine-tuned, English occurrence=language confusion2026.04 | 96.44 | 99.9 | |
| DPOMode=Fine-tuned, English occurrence=language confusion2026.04 | 87.91 | 95.92 | |
| ORPOMode=Fine-tuned, English occurrence=language confusion2026.04 | 83.72 | 96.26 | |
| BaselineDescription=Original model prior to fine-tuning2026.04 | 83.48 | 99.65 | |
| SFTMode=Fine-tuned, English occurrence=language confusion2026.04 | 62.75 | 98.12 |