SOTAVerified

NetHack

Mean in-game score over 1000 episodes with random seeds not seen during training. See https://arxiv.org/abs/2006.13760 (Section 2.4 Evaluation Protocol) for details.

Papers

Showing 1120 of 28 papers

TitleStatusHype
Motif: Intrinsic Motivation from Artificial Intelligence FeedbackCode1
Selective Perception: Optimizing State Descriptions with Reinforcement Learning for Language Model Actors0
Scaling Laws for Imitation Learning in Single-Agent GamesCode1
LuckyMera: a Modular AI Framework for Building Hybrid NetHack AgentsCode1
Katakomba: Tools and Benchmarks for Data-Driven NetHackCode1
Accelerating exploration and representation learning with offline pre-training0
Dungeons and Data: A Large-Scale NetHack DatasetCode2
Improving Policy Learning via Language Dynamics DistillationCode0
Hierarchical Kickstarting for Skill Transfer in Reinforcement LearningCode1
Insights From the NeurIPS 2021 NetHack ChallengeCode0
Show:102550
← PrevPage 2 of 3Next →

No leaderboard results yet.