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

Environments

Benchmarks

Task NameDataset NameSOTA ResultTrend
Depth Estimationenvironments (unseen)
DPE1.25
7
Sound Source Localizationenvironments (unseen)
SLE17
7
Depth EstimationEnvironments (seen)
DPE164
6
Sound Source LocalizationEnvironments (seen)
SLE14.9
6
Multi-step rollout prediction7 Environments (AlfWorld, BabyAI, Maze, SciWorld, TextCraft, WebShop, Wordle) (held-out episodes)
Token F1 (t=1)69
5
Neural Forward Frame RenderingSix Training Environments (val)
PSNR20.96
2
Showing 6 of 6 rows