Science QA with Short Final Answers on SCI-QA
80.2AccuracyCLSR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CLSRLLM=QWEN3-32B2026.06 | 80.2 | 39 | |
| SoTLLM=QWEN3-8B2026.06 | 79.3 | 45 | |
| Raw CoTLLM=QWEN3-32B2026.06 | 79.2 | 81 | |
| CCoTLLM=QWEN3-32B2026.06 | 79.1 | 56 | |
| SoTLLM=QWEN3-32B2026.06 | 78.9 | 48 | |
| CCoTLLM=QWEN3-8B2026.06 | 78.2 | 49 | |
| Raw CoTLLM=QWEN3-8B2026.06 | 77.8 | 75 | |
| CLSRLLM=QWEN3-8B2026.06 | 77.8 | 35 | |
| CoDLLM=QWEN3-32B2026.06 | 77 | 53 | |
| CoDLLM=QWEN3-8B2026.06 | 76.2 | 46 | |
| CLSRLLM=DEEPSEEK-R1-0528-QWEN3-8B2026.06 | 71.5 | 30 | |
| Raw CoTLLM=DEEPSEEK-R1-0528-QWEN3-8B2026.06 | 71.3 | 125 | |
| SoTLLM=DEEPSEEK-R1-0528-QWEN3-8B2026.06 | 71 | 41 | |
| CCoTLLM=DEEPSEEK-R1-0528-QWEN3-8B2026.06 | 70.2 | 43 | |
| CoDLLM=DEEPSEEK-R1-0528-QWEN3-8B2026.06 | 68.4 | 37 | |
| CLSRLLM=LLaMA3-8B2026.06 | 61.4 | 18 | |
| Raw CoTLLM=LLaMA3-8B2026.06 | 61.3 | 51 | |
| SoTLLM=LLaMA3-8B2026.06 | 61.2 | 19 | |
| CCoTLLM=LLaMA3-8B2026.06 | 60.2 | 22 | |
| CoDLLM=LLaMA3-8B2026.06 | 59.8 | 19 |