The PokeAgent Challenge: Competitive and Long-Context Learning at Scale
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908893068656640 |
|---|---|
| author | Karten, Seth Grigsby, Jake Upaa Jr, Tersoo Bae, Junik Hong, Seonghun Jeong, Hyunyoung Jung, Jaeyoon Kerdthaisong, Kun Kim, Gyungbo Kim, Hyeokgi Kim, Yujin Kwon, Eunju Liu, Dongyu Mariglia, Patrick Park, Sangyeon Schink, Benedikt Shi, Xianwei Sistilli, Anthony Twin, Joseph Urdu, Arian Urdu, Matin Wang, Qiao Wu, Ling Zhang, Wenli Zhou, Kunsheng Milani, Stephanie Vodrahalli, Kiran Zhang, Amy Fang, Fei Zhu, Yuke Jin, Chi |
| author_facet | Karten, Seth Grigsby, Jake Upaa Jr, Tersoo Bae, Junik Hong, Seonghun Jeong, Hyunyoung Jung, Jaeyoon Kerdthaisong, Kun Kim, Gyungbo Kim, Hyeokgi Kim, Yujin Kwon, Eunju Liu, Dongyu Mariglia, Patrick Park, Sangyeon Schink, Benedikt Shi, Xianwei Sistilli, Anthony Twin, Joseph Urdu, Arian Urdu, Matin Wang, Qiao Wu, Ling Zhang, Wenli Zhou, Kunsheng Milani, Stephanie Vodrahalli, Kiran Zhang, Amy Fang, Fei Zhu, Yuke Jin, Chi |
| contents | We present the PokeAgent Challenge, a large-scale benchmark for decision-making research built on Pokemon's multi-agent battle system and expansive role-playing game (RPG) environment. Partial observability, game-theoretic reasoning, and long-horizon planning remain open problems for frontier AI, yet few benchmarks stress all three simultaneously under realistic conditions. PokeAgent targets these limitations at scale through two complementary tracks: our Battling Track, which calls for strategic reasoning and generalization under partial observability in competitive Pokemon battles, and our Speedrunning Track, which requires long-horizon planning and sequential decision-making in the Pokemon RPG. Our Battling Track supplies a dataset of 20M+ battle trajectories alongside a suite of heuristic, RL, and LLM-based baselines capable of high-level competitive play. Our Speedrunning Track provides the first standardized evaluation framework for RPG speedrunning, including an open-source multi-agent orchestration system for modular, reproducible comparisons of harness-based LLM approaches. Our NeurIPS 2025 competition validates both the quality of our resources and the research community's interest in Pokemon, with over 100 teams competing across both tracks and winning solutions detailed in our paper. Participant submissions and our baselines reveal considerable gaps between generalist (LLM), specialist (RL), and elite human performance. Analysis against the BenchPress evaluation matrix shows that Pokemon battling is nearly orthogonal to standard LLM benchmarks, measuring capabilities not captured by existing suites and positioning Pokemon as an unsolved benchmark that can drive RL and LLM research forward. We transition to a living benchmark with a live leaderboard for Battling and self-contained evaluation for Speedrunning at https://pokeagentchallenge.com. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_15563 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | The PokeAgent Challenge: Competitive and Long-Context Learning at Scale Karten, Seth Grigsby, Jake Upaa Jr, Tersoo Bae, Junik Hong, Seonghun Jeong, Hyunyoung Jung, Jaeyoon Kerdthaisong, Kun Kim, Gyungbo Kim, Hyeokgi Kim, Yujin Kwon, Eunju Liu, Dongyu Mariglia, Patrick Park, Sangyeon Schink, Benedikt Shi, Xianwei Sistilli, Anthony Twin, Joseph Urdu, Arian Urdu, Matin Wang, Qiao Wu, Ling Zhang, Wenli Zhou, Kunsheng Milani, Stephanie Vodrahalli, Kiran Zhang, Amy Fang, Fei Zhu, Yuke Jin, Chi Machine Learning Artificial Intelligence We present the PokeAgent Challenge, a large-scale benchmark for decision-making research built on Pokemon's multi-agent battle system and expansive role-playing game (RPG) environment. Partial observability, game-theoretic reasoning, and long-horizon planning remain open problems for frontier AI, yet few benchmarks stress all three simultaneously under realistic conditions. PokeAgent targets these limitations at scale through two complementary tracks: our Battling Track, which calls for strategic reasoning and generalization under partial observability in competitive Pokemon battles, and our Speedrunning Track, which requires long-horizon planning and sequential decision-making in the Pokemon RPG. Our Battling Track supplies a dataset of 20M+ battle trajectories alongside a suite of heuristic, RL, and LLM-based baselines capable of high-level competitive play. Our Speedrunning Track provides the first standardized evaluation framework for RPG speedrunning, including an open-source multi-agent orchestration system for modular, reproducible comparisons of harness-based LLM approaches. Our NeurIPS 2025 competition validates both the quality of our resources and the research community's interest in Pokemon, with over 100 teams competing across both tracks and winning solutions detailed in our paper. Participant submissions and our baselines reveal considerable gaps between generalist (LLM), specialist (RL), and elite human performance. Analysis against the BenchPress evaluation matrix shows that Pokemon battling is nearly orthogonal to standard LLM benchmarks, measuring capabilities not captured by existing suites and positioning Pokemon as an unsolved benchmark that can drive RL and LLM research forward. We transition to a living benchmark with a live leaderboard for Battling and self-contained evaluation for Speedrunning at https://pokeagentchallenge.com. |
| title | The PokeAgent Challenge: Competitive and Long-Context Learning at Scale |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.15563 |