Table Chain of Thought Reasoning on TableBench
54.28RgeDeepSeek-V3
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepSeek-V3Model Scale Group=Larger2025.12 | 54.28 | |
| Qwen-PlusModel Scale Group=Larger2025.12 | 52.07 | |
| Qwen3-32BModel Scale Group=Larger2025.12 | 51.74 | |
| QwQ-32BModel Scale Group=Larger2025.12 | 48.83 | |
| TableGPT-R1-8BModel Scale Group=Comparable2025.12 | 48.28 | |
| Qwen3-72BModel Scale Group=Larger2025.12 | 47.63 | |
| TableGPT2-7BModel Scale Group=Comparable2025.12 | 46.19 | |
| Qwen3-14BModel Scale Group=Larger2025.12 | 46.07 | |
| GPT-4oModel Scale Group=Larger2025.12 | 45.71 | |
| Qwen3-8BModel Scale Group=Comparable2025.12 | 41.71 | |
| Table-R1-Zero-7BModel Scale Group=Comparable2025.12 | 39.52 | |
| Llama-3.1-8BModel Scale Group=Comparable2025.12 | 15.26 | |
| TableLLMModel Scale Group=Comparable2025.12 | 3.18 |