Science Question Answering on SciCU
97.1AccuracyKimi-K2.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-K2.5Base Model=Kimi-K2.52026.05 | 97.1 | |
| Qwen3.5-397BBase Model=Qwen3.5-397B2026.05 | 96.2 | |
| GLM-5.1Base Model=GLM-5.12026.05 | 95.9 | |
| Deepseek-V3.2Base Model=Deepseek-V3.22026.05 | 95.6 | |
| Qwen3.5-9B + MaRBase Model=Qwen3.5-9B, Optimization Method=MaR2026.05 | 95.5 | |
| GPT-OSS-120BBase Model=GPT-OSS-120B2026.05 | 94.9 | |
| Qwen3.5-122BBase Model=Qwen3.5-122B2026.05 | 94.9 | |
| Qwen3.5-4B + MaRBase Model=Qwen3.5-4B, Optimization Method=MaR2026.05 | 94.9 | |
| Qwen3.5-35BBase Model=Qwen3.5-35B2026.05 | 94.7 | |
| Qwen3.5-9B + DAPOBase Model=Qwen3.5-9B, Optimization Method=DAPO2026.05 | 94.4 | |
| Qwen3.5-9BBase Model=Qwen3.5-9B2026.05 | 93.8 | |
| Qwen3.5-4B + DAPOBase Model=Qwen3.5-4B, Optimization Method=DAPO2026.05 | 93.5 | |
| Qwen3.5-4BBase Model=Qwen3.5-4B2026.05 | 93 |