#artificial intelligence
May 2026
PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?
PTCG-Bench is presented, a benchmark built on the Pok'e}mon Trading Card Game (PTCG) that evaluates LLM agents at two complementary levels: their decision-making performance within a single complex environment, and their ability to self-evolving through accumulated experience.
Dong-Dong Hua, Yifei Sun, Renhong Huang et al.
· arXiv.org · 0 citations