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

Jericho

Benchmarks

Task NameDataset NameSOTA ResultTrend
Interactive FictionJericho OOD (held-out games)
Average Score (Balances)11.2
9
Sequential Decision-MakingJericho ID (meta-train)
Score (Detective)270.5
9
Text-based game playingJericho Zork3 (test)
Avg Score3.1
7
Text-based game playingJericho Zork1 (test)
Average Score53
7
Text-based game playingJericho Library (test)
Average Score25.9
7
Text-based Reinforcement LearningJericho benchmark (test)
Zork1 (Eps)35
6
Interactive FictionJericho (test)
Zork1 Completion Score40
4
Text-Based AdventureJericho
Omniquest Score10
3
Showing 8 of 8 rows