# Grounding Natural Language Commands to StarCraft II Game States for Narration-Guided Reinforcement Learning

arXiv: https://arxiv.org/abs/1906.02671 · HTML: https://mlsanity.com/paper/1906.02671/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Apr 2019.

## Built on by (most cited first, 5 of 5)

- [Language Grounded Multi-agent Reinforcement Learning with Human-interpretable Communication](https://mlsanity.com/paper/2409.17348/) (2024)
- [Interactive Hierarchical Guidance using Language](https://mlsanity.com/paper/2110.04649/) (2021)
- [On games and simulators as a platform for development of artificial intelligence for command and control](https://mlsanity.com/paper/2110.11305/) (2021)
- [Adaptive Command: Real-Time Policy Adjustment via Language Models in StarCraft II](https://mlsanity.com/paper/2508.16580/) (2025)
- [Improve the efficiency of deep reinforcement learning through semantic exploration guided by natural language](https://mlsanity.com/paper/2309.11753/) (2023)

## Abstract (arXiv)

While deep reinforcement learning techniques have led to agents that are successfully able to learn to perform a number of tasks that had been previously unlearnable, these techniques are still susceptible to the longstanding problem of em reward sparsity. This is especially true for tasks such as training an agent to play StarCraft II, a real-time strategy game where reward is only given at the end of a game which is usually very long. While this problem can be addressed through reward shaping, such approaches typically require a human expert with specialized knowledge. Inspired by the vision of enabling reward shaping through the more-accessible paradigm of natural-language narration, we investigate to what extent we can contextualize these narrations by grounding them to the goal-specific states. We present a mutual-embedding model using a multi-input deep-neural network that projects a sequence of natural language commands into the same high-dimensional representation space as corresponding goal states. We show that using this model we can learn an embedding space with separable and distinct clusters that accurately maps natural-language commands to corresponding game states . We also discuss how this model can allow for the use of narrations as a robust form of reward shaping to improve RL performance and efficiency.