Table Question Answering on MMQA
86.08AccuracyGPT-5-nano
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5-nanoWorkflow=Tree-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 86.08 | |
| GPT-oss-20bWorkflow=Chain-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 83.54 | |
| GPT-5-nanoWorkflow=Chain-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 83.54 | |
| GPT-oss-20bWorkflow=Tree-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 79.75 | |
| GPT-4.1-nanoWorkflow=Tree-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 73.42 | |
| GPT-4.1-nanoWorkflow=Chain-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 67.09 | |
| Ministral-3-8BWorkflow=Tree-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 56.96 | |
| Ministral-3-8BWorkflow=Chain-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 49.37 | |
| QWEN3-8BWorkflow=Tree-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 44.94 | |
| QWEN3-8BWorkflow=Chain-of-Table (Tools), ST Reward=true, SR Reward=RG (TABROUGE)2026.01 | 44.3 |