Free-form QA on FetaQA
40.5BLEUSpecialist SOTA
Evaluation results
| Method | Links | |
|---|---|---|
| Specialist SOTASetting=Others2025.11 | 40.5 | |
| TAMO+Setting=Tuned LLM (SFT)2025.11 | 39.01 | |
| TableLlamaSetting=Tuned LLM (SFT)2025.11 | 38.12 | |
| TAMOSetting=Frozen LLM, Backbone=Llama 3.1 8B2025.11 | 38.09 | |
| TAMO+Setting=Tuned LLM (LoRA)2025.11 | 37.43 | |
| SFTSetting=Tuned LLM (SFT)2025.11 | 37.37 | |
| GPT-4.1Setting=Others2025.11 | 36.75 | |
| TAMOSetting=Frozen LLM2025.11 | 36.52 | |
| Prompt tuningSetting=Frozen LLM, Backbone=Llama 3.1 8B2025.11 | 36.16 | |
| LoRASetting=Tuned LLM (LoRA)2025.11 | 35.8 | |
| Prompt tuningSetting=Frozen LLM2025.11 | 29.94 | |
| GPT-3.5Setting=Others2025.11 | 26.49 | |
| GPT-4Setting=Others2025.11 | 21.7 | |
| Zero-shotSetting=Inference Only2025.11 | 20.08 | |
| Llama 3.1 8BSetting=Inference Only2025.11 | 14.05 | |
| DeepSeek-R1Setting=Others2025.11 | 13.1 |