| Benchmark | Model | Rank | Results |
|---|---|---|---|
| question-answering-on-boolq | AMD OLMo SFT 1B + Recover-LoRA | #88 | Accuracy: 62.17 |
| question-answering-on-piqa | DeepSeek-R1-Distill-Qwen-1.5B + Recover-LoRA | #177 | Accuracy: 54.79 |
| sentence-completion-on-hellaswag | DeepSeek-R1-Distill-Qwen-1.5B + Recover-LoRA | #220 | Accuracy: 26.52 |