Chart Understanding on CharXiv RQ
77.9Pass@1Kimi K2.5 + ETCHR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Kimi K2.5 + ETCHRTemperature=02026.05 | 77.9 | — | |
| Kimi K2.5Temperature=02026.05 | 77.1 | — | |
| Gemini-3.1-Flash-Lite + ETCHRTemperature=02026.05 | 70.8 | — | |
| Gemini-3.1-Flash-LiteTemperature=02026.05 | 68.6 | — | |
| Qwen3-VL-8B + ETCHRTemperature=02026.05 | 49.7 | — | |
| DeepEyesV2Temperature=0, Max tool invocations=102026.05 | 48.9 | — | |
| Qwen3-VL-8BTemperature=02026.05 | 47.9 | — | |
| ThymeTemperature=0, Max tool invocations=102026.05 | 45.3 | — | |
| Bagel-Zebra-CoTTemperature=0, Max image generations=102026.05 | 37.4 | — | |
| ThinkMorph-7BTemperature=0, Max image generations=102026.05 | 28.1 | — | |
| Qwen3-VL-8BFine-tuning setting=Base2026.07 | — | 32.7 | |
| Qwen3-VL-8BFine-tuning setting=FFT2026.07 | — | 23.4 | |
| Qwen3-VL-8BFine-tuning setting=LoRA2026.07 | — | 36.7 | |
| Qwen3-VL-8BFine-tuning setting=GRPO2026.07 | — | 28.8 | |
| Qwen3-VL-8BFine-tuning setting=GSPO2026.07 | — | 30 | |
| Qwen3-VL-8BFine-tuning setting=CPO2026.07 | — | 44.2 |