TacticCraft: Natural Language-Driven Tactical Adaptation for StarCraft II
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909697900019712 |
|---|---|
| author | Ma, Weiyu Jiang, Jiwen Fu, Haobo Zhang, Haifeng |
| author_facet | Ma, Weiyu Jiang, Jiwen Fu, Haobo Zhang, Haifeng |
| contents | We present an adapter-based approach for tactical conditioning of StarCraft II AI agents. Current agents, while powerful, lack the ability to adapt their strategies based on high-level tactical directives. Our method freezes a pre-trained policy network (DI-Star) and attaches lightweight adapter modules to each action head, conditioned on a tactical tensor that encodes strategic preferences. By training these adapters with KL divergence constraints, we ensure the policy maintains core competencies while exhibiting tactical variations. Experimental results show our approach successfully modulates agent behavior across tactical dimensions including aggression, expansion patterns, and technology preferences, while maintaining competitive performance. Our method enables flexible tactical control with minimal computational overhead, offering practical strategy customization for complex real-time strategy games. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_15618 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TacticCraft: Natural Language-Driven Tactical Adaptation for StarCraft II Ma, Weiyu Jiang, Jiwen Fu, Haobo Zhang, Haifeng Artificial Intelligence We present an adapter-based approach for tactical conditioning of StarCraft II AI agents. Current agents, while powerful, lack the ability to adapt their strategies based on high-level tactical directives. Our method freezes a pre-trained policy network (DI-Star) and attaches lightweight adapter modules to each action head, conditioned on a tactical tensor that encodes strategic preferences. By training these adapters with KL divergence constraints, we ensure the policy maintains core competencies while exhibiting tactical variations. Experimental results show our approach successfully modulates agent behavior across tactical dimensions including aggression, expansion patterns, and technology preferences, while maintaining competitive performance. Our method enables flexible tactical control with minimal computational overhead, offering practical strategy customization for complex real-time strategy games. |
| title | TacticCraft: Natural Language-Driven Tactical Adaptation for StarCraft II |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2507.15618 |