🧪 Test?View on arXiv
SpireBench: Evaluating Long-Horizon Strategic Play of Agentic Models with Slay the Spire
Not provided
reasoningbenchmarkingstrategic decision-making
2610.12596
Builder Relevance
8h ago70%
Abstract
SpireBench introduces a benchmark for evaluating long-horizon strategic reasoning in agentic large language models using the game Slay the Spire.
Reality Card
Core Claim
SpireBench demonstrates that while frontier models can match average human players in Slay the Spire, they significantly lag behind expert players, particularly in long-term strategic decision-making.
Method / Result
20 large language models were tested, with frontier models matching average human performance but falling short against experts as run length increased.
Limitations
Models do not improve from the memories they write across runs, often leading to incorrect and overly cautious decisions.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.