Factuality Evaluation on HaluEval QA (500 items)
86.4MC1 ScoreQwen3-30B-A3B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-30B-A3BI(M)=3.51, Branch=mix2026.05 | 86.4 | 71.98 | |
| Ministral-3-14B-RI(M)=0.57, Branch=early2026.05 | 83 | 67.12 | |
| Qwen3-14BI(M)=2.57, Branch=mix2026.05 | 79.4 | 70.4 | |
| LLaMA-30BI(M)=0.13, Branch=early2026.05 | 78.8 | 61.53 | |
| DeepSeek-R1-Q-32BI(M)=1.18, Branch=mix2026.05 | 76.4 | 63.68 | |
| LLaMA-7BI(M)=0.21, Branch=early2026.05 | 75.8 | 58.8 | |
| LLaMA-13BI(M)=0.16, Branch=early2026.05 | 75.6 | 59 | |
| GPT-OSS-20BI(M)=3.48, Branch=mix2026.05 | 67 | 59.32 | |
| Llama-3.3-70BI(M)=0.39, Branch=early2026.05 | 61 | 52.3 | |
| Phi-4-ReasoningI(M)=0.27, Branch=early2026.05 | 57.4 | 50.23 | |
| Mixtral-8×7BI(M)=1.30, Branch=mix2026.05 | 55.4 | 47.81 | |
| GPT-OSS-120BI(M)=5.66, Branch=mix2026.05 | 49.2 | 47.79 | |
| Gemma-3-4BI(M)=11.91, Branch=mix2026.05 | 48.2 | 43.99 | |
| Gemma-3-27BI(M)=5.18, Branch=mix2026.05 | 47 | 44.13 | |
| TRACEBase Model=Gemma-3-1B, I(M)=6.54, Branch=mix2026.05 | 43.6 | 43.38 | |
| Gemma-3-1BI(M)=6.54, Branch=mix2026.05 | 41.2 | 39.13 | |
| TRACEBase Model=Gemma-3-4B, I(M)=11.91, Branch=mix2026.05 | 34.8 | 37.82 | |
| TRACEBase Model=Gemma-3-27B, I(M)=5.18, Branch=mix2026.05 | 24.8 | 26.97 | |
| TRACEBase Model=GPT-OSS-120B, I(M)=5.66, Branch=mix2026.05 | 15.4 | 12.33 | |
| TRACEBase Model=Mixtral-8×7B, I(M)=1.30, Branch=mix2026.05 | 14.4 | 11.29 | |
| TRACEBase Model=Phi-4-Reasoning, I(M)=0.27, Branch=early2026.05 | 14 | 12.06 | |
| TRACEBase Model=GPT-OSS-20B, I(M)=3.48, Branch=mix2026.05 | 12.8 | 13.75 | |
| TRACEBase Model=Llama-3.3-70B, I(M)=0.39, Branch=early2026.05 | 8.4 | 7.15 | |
| TRACEBase Model=Ministral-3-14B-R, I(M)=0.57, Branch=early2026.05 | 5.8 | 3.11 | |
| TRACEBase Model=DeepSeek-R1-Q-32B, I(M)=1.18, Branch=mix2026.05 | 5.4 | 8.06 | |
| TRACEBase Model=LLaMA-30B, I(M)=0.13, Branch=early2026.05 | 5.2 | 4.15 | |
| TRACEBase Model=LLaMA-13B, I(M)=0.16, Branch=early2026.05 | 5 | 4.67 | |
| TRACEBase Model=Qwen3-14B, I(M)=2.57, Branch=mix2026.05 | 3.4 | 5.3 | |
| TRACEBase Model=LLaMA-7B, I(M)=0.21, Branch=early2026.05 | 2.4 | 2.62 | |
| TRACEBase Model=Qwen3-30B-A3B, I(M)=3.51, Branch=mix2026.05 | 1 | 6.99 |