Question Answering on ARC Challenge (ARC-C)
64.1ARC-C ScoreMARI
Evaluation Results
| Method | Links | |
|---|---|---|
| MARIModel=Qwen2.5-14B2026.05 | 64.1 | |
| ITIModel=Qwen2.5-14B2026.05 | 63.8 | |
| ReFTModel=Qwen2.5-14B2026.05 | 63.6 | |
| VanillaModel=Qwen2.5-14B2026.05 | 63.4 | |
| NL-ITIModel=Qwen2.5-14B2026.05 | 63.3 | |
| CAAModel=Qwen2.5-14B2026.05 | 63.1 | |
| MARIModel=Qwen2.5-32B2026.05 | 60 | |
| ITIModel=Qwen2.5-32B2026.05 | 59.6 | |
| ReFTModel=Qwen2.5-32B2026.05 | 59.5 | |
| VanillaModel=Qwen2.5-32B2026.05 | 59.3 | |
| NL-ITIModel=Qwen2.5-32B2026.05 | 59.2 | |
| CAAModel=Qwen2.5-32B2026.05 | 59 | |
| MARIModel=Qwen2-7B2026.05 | 52.3 | |
| ITIModel=Qwen2-7B2026.05 | 52.2 | |
| MARIModel=Llama-3-8B2026.05 | 52.1 | |
| ReFTModel=Qwen2-7B2026.05 | 52.1 | |
| VanillaModel=Qwen2-7B2026.05 | 51.9 | |
| NL-ITIModel=Qwen2-7B2026.05 | 51.8 | |
| ITIModel=Llama-3-8B2026.05 | 51.7 | |
| ReFTModel=Llama-3-8B2026.05 | 51.6 | |
| CAAModel=Qwen2-7B2026.05 | 51.6 | |
| VanillaModel=Llama-3-8B2026.05 | 51.4 | |
| NL-ITIModel=Llama-3-8B2026.05 | 51.3 | |
| CAAModel=Llama-3-8B2026.05 | 51.1 | |
| ITIModel=Llama-2-13B2026.05 | 42.9 | |
| ReFTModel=Llama-2-13B2026.05 | 42.8 | |
| VanillaModel=Llama-2-13B2026.05 | 42.6 | |
| NL-ITIModel=Llama-2-13B2026.05 | 42.5 | |
| MARIModel=Llama-2-13B2026.05 | 42.4 | |
| CAAModel=Llama-2-13B2026.05 | 42.3 | |
| ITIModel=Llama-2-7B2026.05 | 34.1 | |
| ReFTModel=Llama-2-7B2026.05 | 34 | |
| VanillaModel=Llama-2-7B2026.05 | 33.8 | |
| NL-ITIModel=Llama-2-7B2026.05 | 33.7 | |
| CAAModel=Llama-2-7B2026.05 | 33.5 | |
| MARIModel=Llama-2-7B2026.05 | 33.5 |