Model Card Generation on MetaGAI-Bench Model Cards
4Faithfulness (G1) RankZero-shot
Evaluation Results
| Method | Links | ||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Zero-shotCard Type=Model2026.03 | 4 | 2.73 | 4 | 2.94 | 4 | 4.12 | 4 | 3.01 | 4 | 2.69 | 4 | 2.85 | 4 | 3.7 | 4 | 3.17 | 4 | 2.68 | 4 | 2.79 | 4 | 3.18 | 4 | 2.95 | |
| Zero-shotLLM Judge=GPT-5-nano2026.03 | 4 | — | — | — | 4 | — | — | — | 3 | — | — | — | 1 | — | — | — | 4 | — | — | — | 3.2 | — | — | — | |
| Human-AnnotatedLLM Judge=Gemini-2.5 Flash-Lite2026.03 | 4 | — | — | — | 4 | — | — | — | 4 | — | — | — | 4 | — | — | — | 4 | — | — | — | 4 | — | — | — | |
| CardGenCard Type=Model2026.03 | 3 | 3.29 | 3 | 3.21 | 2 | 4.53 | 3 | 3.33 | 3 | 3.28 | 3 | 3.17 | 2 | 4.19 | 3 | 3.39 | 3 | 3.26 | 3 | 3.09 | 2.6 | 3.71 | 3 | 3.24 | |
| Zero-shotLLM Judge=Gemini-2.5 Flash-Lite2026.03 | 3 | — | — | — | 3 | — | — | — | 3 | — | — | — | 1 | — | — | — | 3 | — | — | — | 2.6 | — | — | — | |
| CardGenLLM Judge=GPT-5-nano2026.03 | 3 | — | — | — | 2 | — | — | — | 2 | — | — | — | 3 | — | — | — | 2 | — | — | — | 2.4 | — | — | — | |
| AdaQE-CGCard Type=Model2026.03 | 2 | 3.47 | 2 | 3.31 | 3 | 4.38 | 2 | 3.43 | 2 | 3.44 | 2 | 3.22 | 3 | 4.05 | 2 | 3.45 | 2 | 3.35 | 2 | 3.17 | 2.4 | 3.74 | 2 | 3.32 | |
| CardGenLLM Judge=Gemini-2.5 Flash-Lite2026.03 | 2 | — | — | — | 2 | — | — | — | 2 | — | — | — | 3 | — | — | — | 2 | — | — | — | 2.2 | — | — | — | |
| AdaQE-CGLLM Judge=GPT-5-nano2026.03 | 2 | — | — | — | 1 | — | — | — | 4 | — | — | — | 4 | — | — | — | 1 | — | — | — | 2.4 | — | — | — | |
| Human-AnnotatedCard Type=Model2026.03 | 1 | 3.69 | 1 | 3.53 | 1 | 4.66 | 1 | 3.62 | 1 | 3.66 | 1 | 3.43 | 1 | 4.32 | 1 | 3.59 | 1 | 3.55 | 1 | 3.35 | 1 | 3.98 | 1 | 3.5 | |
| Human-AnnotatedLLM Judge=GPT-5-nano2026.03 | 1 | — | — | — | 3 | — | — | — | 1 | — | — | — | 2 | — | — | — | 3 | — | — | — | 2 | — | — | — | |
| AdaQE-CGLLM Judge=Gemini-2.5 Flash-Lite2026.03 | 1 | — | — | — | 1 | — | — | — | 1 | — | — | — | 2 | — | — | — | 1 | — | — | — | 1.2 | — | — | — |