Table Question Answering on FeTaQA (test) benchmark leaderboard · SOTA2 Research

Table Question Answering on FeTaQA (test)

19.32sacreBLEUMeta-Llama-3-8B-Instruct

Evaluation Results

Method	Links
Meta-Llama-3-8B-Instruct2026.05		19.32	0.58	0.35	0.47	0.48	0.56	0.6
Meta-Llama-3-8B-Instruct2026.05		17.96	0.54	0.33	0.45	0.45	0.52	0.58
DeepSeek-R1-Distill-Llama-8B2026.05		17.64	0.54	0.32	0.44	0.45	0.52	0.58
Llama-3.1-8B-Instruct2026.05		17.28	0.54	0.32	0.44	0.44	0.51	0.58
Meta-Llama-3-8B-Instruct2026.05		17.01	0.52	0.31	0.43	0.42	0.67	0.54
TableGPT2-7B2026.05		15.37	0.48	0.29	0.39	0.4	0.44	0.58
Llama-3.1-8B-Instruct2026.05		14.83	0.47	0.28	0.39	0.38	0.42	0.51
DeepSeek-R1-Distill-Llama-8B2026.05		13.91	0.47	0.27	0.38	0.38	0.44	0.51