Chart Question Answering on PolyChartQA MLM-generated questions
86.94AccuracyGPT-4.1
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4.1Prompting=CoT2026.04 | 86.94 | |
| GPT-4.1Prompting=Zero-shot2026.04 | 85.98 | |
| Gemini-2.0-flashPrompting=CoT2026.04 | 81.06 | |
| Gemini-2.0-flashPrompting=Zero-shot2026.04 | 80.28 | |
| Claude-3.7-SonnetPrompting=Zero-shot2026.04 | 74.94 | |
| Claude-3.7-SonnetPrompting=CoT2026.04 | 74.71 | |
| Pixtral-12BPrompting=Zero-shot2026.04 | 63.77 | |
| Pixtral-12BPrompting=CoT2026.04 | 63.54 | |
| ChartGemmaPrompting=Zero-shot2026.04 | 28.28 | |
| Gemma-3-4b-itPrompting=Zero-shot2026.04 | 27.49 | |
| Gemma-3-4b-itPrompting=CoT2026.04 | 25.29 | |
| ChartGemmaPrompting=CoT2026.04 | 23.36 | |
| LLaMA-3.2-11B-VisionPrompting=Zero-shot2026.04 | 17.94 | |
| LLaMA-3.2-11B-VisionPrompting=CoT2026.04 | 15.59 | |
| LLaVA1.5-7bPrompting=Zero-shot2026.04 | 11.68 | |
| LLaVA1.5-7bPrompting=CoT2026.04 | 11.59 | |
| MatChaPrompting=Zero-shot2026.04 | 5.43 | |
| MatChaPrompting=CoT2026.04 | 4.41 |