Multi-chart Question Answering on PolyChartQA (human-authored)
71.87H-AccuracyGemini-2.0-flash
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.0-flashSetting=CoT2026.04 | 71.87 | |
| Gemini-2.0-flashSetting=Zero-shot2026.04 | 70.52 | |
| Claude-3.7-SonnetSetting=CoT2026.04 | 66.09 | |
| Claude-3.7-SonnetSetting=Zero-shot2026.04 | 65.32 | |
| GPT-4.1Setting=Zero-shot2026.04 | 61.08 | |
| GPT-4.1Setting=CoT2026.04 | 59.54 | |
| Pixtral-12BSetting=CoT2026.04 | 53.76 | |
| Pixtral-12BSetting=Zero-shot2026.04 | 53.18 | |
| ChartGemmaSetting=CoT2026.04 | 23.51 | |
| Gemma-3-4b-itSetting=CoT2026.04 | 22.74 | |
| Gemma-3-4b-itSetting=Zero-shot2026.04 | 21.19 | |
| ChartGemmaSetting=Zero-shot2026.04 | 19.27 | |
| LLaMA-3.2-11BSetting=Zero-shot2026.04 | 15.22 | |
| LLaVA-1.5-7BSetting=Zero-shot2026.04 | 12.33 | |
| LLaVA-1.5-7BSetting=CoT2026.04 | 11.75 | |
| MatChaSetting=Zero-shot2026.04 | 9.06 | |
| LLaMA-3.2-11BSetting=CoT2026.04 | 8.09 | |
| MatChaSetting=CoT2026.04 | 6.17 |