Table Question Answering on MultiTabQA geoQuery
49.36PrecisionTAMO
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TAMOSetting=Tuned LLM (SFT), Variant=TAMO+SFT2025.11 | 49.36 | 25.46 | 33.59 | |
| LoRASetting=Tuned LLM (LoRA)2025.11 | 30.56 | 10.3 | 15.41 | |
| SFTSetting=Tuned LLM (SFT)2025.11 | 30.55 | 11.04 | 16.22 | |
| TAMOSetting=Tuned LLM (LoRA), Variant=TAMO+LoRA2025.11 | 28.32 | 10.67 | 15.5 | |
| One-shotSetting=Inference Only2025.11 | 9.68 | 5.96 | 7.38 | |
| TAMOSetting=Frozen LLM2025.11 | 6.82 | 4.86 | 5.67 | |
| Prompt tuningSetting=Frozen LLM2025.11 | 4.83 | 3.46 | 4.03 |