mud
PulseAugur coverage of mud — every cluster mentioning mud across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
AI agents tested in retro text-based adventure games
CrucibleBench introduces a novel approach to evaluating AI agents by utilizing Multi-User Dungeons (MUDs) as a testing environment. This method moves away from complex simulators, offering a persistent and constrained s…
-
LLMs evaluated in text-based MUDs to test behavioral adaptation
Researchers have developed CrucibleBench, a novel method for evaluating large language models (LLMs) by placing them within a text-based Multi-User Dungeon (MUD) environment. This approach leverages the inherent constra…
-
New Audiocasts and Gaming Reflections Surface
This cluster aggregates news about recent audiocasts and podcasts, including discussions with Dave Airlie on SE Radio, and new episodes of BSD Now Podcast and This Week in Linux. It also includes a personal reflection o…