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

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–25 of 28 papers

TitleStatusHype
PufferLib: Making Reinforcement Learning Libraries and Environments Play NiceCode4
Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement LearningCode3
Syllabus: Portable Curricula for Reinforcement Learning AgentsCode2
Dungeons and Data: A Large-Scale NetHack DatasetCode2
BeBold: Exploration Beyond the Boundary of Explored RegionsCode1
The NetHack Learning EnvironmentCode1
Hierarchical Kickstarting for Skill Transfer in Reinforcement LearningCode1
Skill Set Optimization: Reinforcing Language Model Behavior via Transferable SkillsCode1
Katakomba: Tools and Benchmarks for Data-Driven NetHackCode1
LuckyMera: a Modular AI Framework for Building Hybrid NetHack AgentsCode1
CORA: Benchmarks, Baselines, and Metrics as a Platform for Continual Reinforcement Learning AgentsCode1
SILG: The Multi-environment Symbolic Interactive Language Grounding BenchmarkCode1
Motif: Intrinsic Motivation from Artificial Intelligence FeedbackCode1
Online Intrinsic Rewards for Decision Making Agents from Large Language Model FeedbackCode1
NovelD: A Simple yet Effective Exploration CriterionCode1
Playing NetHack with LLMs: Potential & Limitations as Zero-Shot AgentsCode1
Scaling Laws for Imitation Learning in Single-Agent GamesCode1
diff History for Neural Language AgentsCode1
MaestroMotif: Skill Design from Artificial Intelligence Feedback—0
BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games—0
Exploration in NetHack With Secret Discovery—0
Accelerating exploration and representation learning with offline pre-training—0
Selective Perception: Optimizing State Descriptions with Reinforcement Learning for Language Model Actors—0
SILG: The Multi-domain Symbolic Interactive Language Grounding Benchmark—0
MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning ResearchCode0
Show:102550
← PrevPage 1 of 2Next →

No leaderboard results yet.