Share your thoughts, 1 month free Claude Pro on usSee more

Table Reasoning on TableBench (DP/CoT Metrics)

44.54DP

Table-R1 (w/ RE-SFT & TARPO)

Updated 20d ago

Evaluation Results

Method	Links
Table-R1 (w/ RE-SFT & TARPO) 2025.05		44.54	49.3	47.21	44.54
QWQ 2025.05		43.87	43.48	37.06	31.58
GPT-4o 2025.05		40.91	51.96	41.43	45.71
Table-R1 (w/ RE-SFT & TARPO) 2025.05		39.29	42.14	38.95	41.6
TableGPT2 2025.05		38.9	50.06	30.47	28.98
Table-R1 (w/ RE-SFT & TARPO) 2025.05		38.5	40.21	38.51	42.78
Table-R1 (w/ RE-SFT & TARPO) 2025.05		36.22	41.85	32.84	41.13
Table-R1 (w/ RE-SFT & TARPO) 2025.05		35.71	39.54	36.78	41.93
RE-TableLLM (w/ RE-SFT) 2025.05		31.53	44.67	34.7	44.36
DS-lite 2025.05		29.6	30.93	22.61	10.9
RE-TableLLM (w/ RE-SFT) 2025.05		29.01	38.57	31.15	41.86
Table-R1 (w/ RE-SFT & TARPO) 2025.05		28.74	35.85	29.62	41.71
RE-TableLLM (w/ RE-SFT) 2025.05		28.71	38.36	30.46	42.14
Qwen2.5-Coder 2025.05		28.67	36.25	25.95	16.15
RE-TableLLM (w/ RE-SFT) 2025.05		28.24	37.33	29.1	41.05
TableGPT2 2025.05		27.95	41.05	31.4	38.67
Table-R1 (w/ RE-SFT & TARPO) 2025.05		26.63	28.42	26.1	41.15
RE-TableLLM (w/ RE-SFT) 2025.05		25.63	31.49	25.88	41.63
Qwen2.5-Instruct 2025.05		25.18	29.77	24.35	22.58
CHAIN-OF-TABLE 2025.05		24.86	32.61	26.39	20.24
TableLLM (w/ SFT) 2025.05		23.15	30.51	23.56	18.74
TableLLM (w/ SFT) 2025.05		22.45	32.65	23.91	25.29
TableLLM (w/ SFT) 2025.05		22.42	35.88	27.58	29.15
RE-TableLLM (w/ RE-SFT) 2025.05		22.38	32.6	25.57	41.64
TableLLM (w/ SFT) 2025.05		22.3	30.77	21.92	27.17
TableLLM (w/ SFT) 2025.05		22.29	31.9	23.62	12.87
Yi-Coder 2025.05		21.94	22.8	8.43	11.36
RE-TableLLM (w/ RE-SFT) 2025.05		21.47	26.49	24.34	42.52
TableLLM (w/ SFT) 2025.05		20.15	24.81	20.55	15.14
TableLLM (w/ SFT) 2025.05		20.08	30.02	22.11	27.55
TableLLM 2025.05		3.88	3.85	2.88	2.94