Hierarchical Table Question Answering on MultiHiertt (test)
70.32AccuracyEEDP
Evaluation Results
| Method | Links | |
|---|---|---|
| EEDPBackbone=GPT-42024.02 | 70.32 | |
| Program-of-Thoughts (PoT)Backbone=GPT-42024.02 | 67.23 | |
| Chain-of-Thought (CoT)Backbone=GPT-42024.02 | 55.13 | |
| Program-of-Thoughts (PoT)Backbone=GPT-3.5-Turbo2024.02 | 52.18 | |
| EEDPBackbone=GPT-3.5-Turbo2024.02 | 49.65 | |
| Chain-of-Thought (CoT)Backbone=GPT-3.5-Turbo2024.02 | 42.33 | |
| DirectBackbone=GPT-42024.02 | 41.35 | |
| EEDPBackbone=PaLM 2-540B2024.02 | 37.97 | |
| DecomposersBackbone=GPT-42024.02 | 36.86 | |
| Program-of-Thoughts (PoT)Backbone=PaLM 2-540B2024.02 | 36.52 | |
| Chain-of-Thought (CoT)Backbone=Llama 2-13B2024.02 | 30.66 | |
| DirectBackbone=GPT-3.5-Turbo2024.02 | 25.88 | |
| EEDPBackbone=Llama 2-13B2024.02 | 24.15 | |
| Chain-of-Thought (CoT)Backbone=Mistral-7B2024.02 | 22.92 | |
| Chain-of-Thought (CoT)Backbone=PaLM 2-540B2024.02 | 20.67 | |
| DecomposersBackbone=GPT-3.5-Turbo2024.02 | 20.61 | |
| DecomposersBackbone=PaLM 2-540B2024.02 | 20.19 | |
| Chain-of-Thought (CoT)Backbone=MAmmoTH-13B2024.02 | 18.56 | |
| EEDPBackbone=MAmmoTH-13B2024.02 | 18.36 | |
| Program-of-Thoughts (PoT)Backbone=Llama 2-13B2024.02 | 18.12 | |
| DecomposersBackbone=Llama 2-13B2024.02 | 16.86 | |
| Program-of-Thoughts (PoT)Backbone=Mistral-7B2024.02 | 14.94 | |
| DirectBackbone=PaLM 2-540B2024.02 | 14.2 | |
| DirectBackbone=Mistral-7B2024.02 | 14.09 | |
| DecomposersBackbone=MAmmoTH-13B2024.02 | 11.87 | |
| DecomposersBackbone=Mistral-7B2024.02 | 11.63 | |
| EEDPBackbone=Mistral-7B2024.02 | 10.97 | |
| DirectBackbone=MAmmoTH-13B2024.02 | 10.12 | |
| Program-of-Thoughts (PoT)Backbone=MAmmoTH-13B2024.02 | 6.57 | |
| DirectBackbone=Llama 2-13B2024.02 | 1.54 |