Metraipolis › archive › dc_aa28f4cafbd3
AceAction-5M: Hierarchical Goals for Generalist Action Learning Across 3D Game Worlds — arXiv:2611.04241
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.04241
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T18:34:29.097Z
- SHA-256 of the archived text
9d8c2ada357b971056821c30779a67d909dc87ceec15c9cb630f4b4e1163cdd8- Stable URL of this page
https://metraipolis.com/documento/dc_aa28f4cafbd3
How to cite
AceAction-5M: Hierarchical Goals for Generalist Action Learning Across 3D Game Worlds — arXiv:2611.04241 — arXiv (en), https://arxiv.org/abs/2611.04241 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aa28f4cafbd3 (Metraipolis archive dc_aa28f4cafbd3, sha256 9d8c2ada357b971056821c30779a67d909dc87ceec15c9cb630f4b4e1163cdd8)
BibTeX
@misc{dcaa28f4cafbd3,
title = {AceAction-5M: Hierarchical Goals for Generalist Action Learning Across 3D Game Worlds — arXiv:2611.04241},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.04241},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aa28f4cafbd3; sha256 9d8c2ada357b971056821c30779a67d909dc87ceec15c9cb630f4b4e1163cdd8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04241. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Video games provide a practical setting for collecting vision-language-action (VLA) data, whose acquisition in physical environments is costly. Existing datasets, however, remain limited by restricted game coverage, sparse language supervision, or temporal hierarchies that do not necessarily reflect changes in task objectives. We introduce AceAction, a dataset of human gameplay across multiple 3D games for learning skill composition and subgoal transitions. AceAction organizes demonstrations around complete tasks, linking overall instructions, ordered subgoals, and local skill instructions to temporally aligned visual observations and keyboard-and-mouse actions. To construct these annotations, we develop a vision-language model pipeline that identifies subgoals from changes in objectives or interaction targets, anchors navigation goals to visually recognizable waypoints, and composes adjacent subgoals into task instructions. The resulting demonstrations cover navigation, reconnaissance, combat, interaction, and recovery, with explicit stage boundaries that connect task progression to fine-grained control. We assess annotation quality through model-based verification and human inspection. An audit of 500 trajectories yields a 95.8% pass rate across independent reviewer assessments. AceAction provides a data foundation for studying how agents execute, combine, and switch skills to follow complex instructions in dynamic game environments. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.04241 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-04T18:32:37.288Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).