Embodied Task Completion on ALFRED unseen (test) benchmark leaderboard · SOTA2 Research

Embodied Task Completion on ALFRED unseen (test)

4,572Success RatePrompter

Evaluation Results

Method	Links
Prompter2023.06		4,572	58.76	2,076	26.22
EPA2023.06		3,607	39.54	292	3.91
LGS-RPA2023.06		3,541	45.24	1,568	22.76
FILM2023.06		2,780	38.52	1,132	15.13
HLSM2023.06		2,446	34.75	967	13.13
ET2023.06		857	18.56	410	11.46
RePlan-Bot2026.05		47.61	60.29	21.89	26.31
Prompter2026.05		45.32	56.57	20.79	25.8
RePlan-Bot2026.05		44.79	57.02	21.89	26.31
CAPEAM2026.05		43.17	56.52	20.13	24.26
Prompter2026.05		41.53	56.57	18.84	25.8
OPEx-S2024.03		41.27	53.82	12.57	18.46
CAPEAM2026.05		40.42	54.17	19.34	23.68
Seq2Seq2023.06		39	7.03	8	4.26
EPA2024.03		36.07	39.54	2.92	3.91
LGS-RPA2026.05		35.41	45.24	22.76	22.76
LGS-RPA2024.03		27.8	38.55	12.92	20.01
LGS-RPA2026.05		27.8	38.55	12.92	20.01
FILM2026.05		27.8	38.52	11.32	15.13
FILM2024.03		24.46	34.75	9.67	13.13
FILM2026.05		24.46	34.75	9.67	13.13
HLSM2024.03		16.29	27.24	4.34	8.45
HLSM2026.05		16.29	27.24	4.34	8.45
ABP2026.05		15.43	24.76	1.08	2.22
LLM-Planner2024.03		13.41	22.89	—	—
LAV2024.03		6.38	17.27	3.12	10.47