mlsanityData to 14 Oct 2026

Complex Instruction Following with Diverse Style Policies in Football Games

arXiv 2511.19885 ↗First posted 25 Nov 2025cs.MA cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL for games and combinatorial optimization

Abstract (arXiv)

Despite advancements in language-controlled reinforcement learning (LC-RL) for basic domains and straightforward commands (e.g., object manipulation and navigation), effectively extending LC-RL to comprehend and execute high-level or abstract instructions in complex, multi-agent environments, such as football games, remains a significant challenge. To address this gap, we introduce Language-Controlled Diverse Style Policies (LCDSP), a novel LC-RL paradigm specifically designed for complex scenarios. LCDSP comprises two key components: a Diverse Style Training (DST) method and a Style Interpreter (SI). The DST method efficiently trains a single policy capable of exhibiting a wide range of diverse behaviors by modulating agent actions through style parameters (SP). The SI is designed to accurately and rapidly translate high-level language instructions into these corresponding SP. Through extensive experiments in a complex 5v5 football environment, we demonstrate that LCDSP effectively comprehends abstract tactical instructions and accurately executes the desired diverse behavioral styles, showcasing its potential for complex, real-world applications.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods. Builds on 25 of its 48 references that are in the map. First posted on arXiv 25 Nov 2025.

Builds on

25 of 48 references.

References · 25

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,348
  4. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,071
  5. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  6. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,103
  7. 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,035
  8. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learningcited by 523
  9. 2021Conflict-Averse Gradient Descent for Multi-task Learningcited by 459
  10. 2023Language to Rewards for Robotic Skill Synthesiscited by 312
  11. 2021A Practical Guide to Multi-Objective Reinforcement Learning and Planningcited by 256
  12. 2019Mastering Complex Control in MOBA Games with Deep Reinforcement Learningcited by 161
13 more
  1. 2019A Survey of Reinforcement Learning Informed by Natural Languagecited by 154
  2. 2020Multi-Task Reinforcement Learning with Soft Modularizationcited by 143
  3. 2019Language as an Abstraction for Hierarchical Deep Reinforcement Learningcited by 138
  4. 2023Large Language Models as Generalizable Policies for Embodied Taskscited by 86
  5. 2016Multi-Objective Deep Reinforcement Learningcited by 83
  6. 2018Learning to Understand Goal Specifications by Modelling Rewardcited by 80
  7. 2019From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Followingcited by 67
  8. 2020Human Instruction-Following with Deep Reinforcement Learning via Transfer-Learning from Textcited by 59
  9. 2018Meta-Learning for Multi-objective Reinforcement Learningcited by 34
  10. 2014Multi-objective Reinforcement Learning with Continuous Pareto Frontier Approximation Supplementary Materialcited by 14
  11. 2023Contrastive Modules with Temporal Attention for Multi-Task Reinforcement Learningcited by 10
  12. 2025Noise-Resilient Symbolic Regression with Dynamic Gating Reinforcement Learningcited by 6
  13. 2023An Empirical Study on Google Research Football Multi-agent Scenarioscited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Policy Gradient Steering: Interventions from Behavioral Objectives
  2. 2021From Motor Control to Team Play in Simulated Humanoid Football
  3. 2025UBCL: A Reinforcement Learning Framework for Controllable and Diverse Player Behaviors
  4. 2025TacEleven: generative tactic discovery for football open play
  5. 2026TacticGen: Grounding Adaptable and Scalable Generation of Football Tactics
  6. 2025Human-Like Goalkeeping in a Realistic Football Simulation: a Sample-Efficient Reinforcement Learning Approach
  7. 2026Coachable agents for interactive gameplay
  8. 2026Synthesizing Reactive Character Behaviors for Continuous Games via Programmatic Policy Search