@討論串

起始文章

AI Models Struggle with Slay the Spire's Highest Difficulty

A recent preprint study by Sander W. Keemink and Scott C. Lowe evaluated 20 large language models (LLMs) playing Slay the Spire at Ascension 20. While some models successfully cleared Act 1, performance dropped significantly in later stages

此文章將加入同一主題的討論串。