Our new X account is live! Follow @wizwand_team for updates

Embodied Task Completion on ALFRED EB

32.2Avg Score

Gemma-3-12B-IT + µOffline Sup.

Updated 4d ago

Evaluation Results

Method	Links
Gemma-3-12B-IT + µOffline Sup. 2026.01		32.2	48	31	32	23	27
Gemma-3-12B-IT + µSimple 2026.01		28	41	27	34	16	22
Gemma-3-12B-IT + µOnline RL 2026.01		27.8	38	29	41	21	26
Gemma-3-12B-IT 2026.01		25.6	32	26	38	20	12
Qwen2.5-VL-7B-Ins + µOnline RL 2026.01		14.2	10	13	21	3	24
Qwen2.5-VL-7B-Ins + µOffline Sup. 2026.01		12.2	9	10	14	2	26
Qwen2.5-VL-7B-Ins + µSimple 2026.01		9.6	15	10	12	2	14
Qwen2.5-VL-7B-Ins 2026.01		4.7	10	8	6	0	2