VLM-based Q/A Quiz on ArcBench VLM-based Q/A Quiz Open
95.69Accuracy (Story)ArcDeck
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ArcDeckGeneration Model=Qwen-3-VL-32B-Instruct2026.04 | 95.69 | 66.58 | 83.22 | 71.39 | |
| ArcDeckGeneration Model=GPT-4o2026.04 | 95.46 | 64.57 | 72.63 | 67.56 | |
| ArcDeckGeneration Model=GPT-52026.04 | 95.41 | 65.95 | 81.13 | 71.58 | |
| SlideGenGeneration Model=GPT-52026.04 | 95.26 | 68.37 | 78.39 | 69.43 | |
| P2PosterGeneration Model=GPT-52026.04 | 95.22 | 65.51 | 76.7 | 67.67 | |
| HTMLGeneration Model=GPT-52026.04 | 95.16 | 64.13 | 81.83 | 69.73 | |
| SlideGenGeneration Model=Qwen-3-VL-32B-Instruct2026.04 | 94.81 | 62.12 | 76.57 | 67.96 | |
| P2PosterGeneration Model=Qwen-3-VL-32B-Instruct2026.04 | 94.28 | 63.36 | 75.96 | 66.75 | |
| P2PosterGeneration Model=GPT-4o2026.04 | 94.12 | 63.11 | 66.07 | 60.18 | |
| HTMLGeneration Model=Qwen-3-VL-32B-Instruct2026.04 | 94.07 | 61.53 | 75.24 | 66.68 | |
| SlideGenGeneration Model=GPT-4o2026.04 | 93.75 | 62.79 | 66.15 | 61 | |
| PPTAgentGeneration Model=GPT-4o2026.04 | 93.35 | 61.9 | 66.01 | 60.84 | |
| HTMLGeneration Model=GPT-4o2026.04 | 93.34 | 57.36 | 64.57 | 60.57 | |
| PPTAgentGeneration Model=GPT-52026.04 | 92.96 | 65.08 | 69.49 | 65.67 | |
| PPTAgentGeneration Model=Qwen-3-VL-32B-Instruct2026.04 | 92.66 | 61.94 | 69.53 | 65.38 |