# Can LLMs Translate Human Instructions into a Reinforcement Learning Agent's Internal Emergent Symbolic Representation?

arXiv: https://arxiv.org/abs/2510.24259 · HTML: https://mlsanity.com/paper/2510.24259/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use.
- Builds on 9 of its 17 references that are in the map.
- First posted on arXiv 28 Oct 2025.

## Placed in

- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 9 of 9)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation](https://mlsanity.com/paper/1609.08144/) (2016)
- [A Comprehensive Overview of Large Language Models](https://mlsanity.com/paper/2307.06435/) (2023)
- [Data-Efficient Hierarchical Reinforcement Learning](https://mlsanity.com/paper/1805.08296/) (2018)
- [Benchmarking Deep Reinforcement Learning for Continuous Control](https://mlsanity.com/paper/1604.06778/) (2016)
- [Symbol Emergence in Cognitive Developmental Systems: a Survey](https://mlsanity.com/paper/1801.08829/) (2018)
- [Goal Space Abstraction in Hierarchical Reinforcement Learning via Set-Based Reachability Analysis](https://mlsanity.com/paper/2309.07675/) (2023)
- [Reconciling Spatial and Temporal Abstractions for Goal Representation](https://mlsanity.com/paper/2401.09870/) (2024)

## Abstract (arXiv)

Emergent symbolic representations are critical for enabling developmental learning agents to plan and generalize across tasks. In this work, we investigate whether large language models (LLMs) can translate human natural language instructions into the internal symbolic representations that emerge during hierarchical reinforcement learning. We apply a structured evaluation framework to measure the translation performance of commonly seen LLMs -- GPT, Claude, Deepseek and Grok -- across different internal symbolic partitions generated by a hierarchical reinforcement learning algorithm in the Ant Maze and Ant Fall environments. Our findings reveal that although LLMs demonstrate some ability to translate natural language into a symbolic representation of the environment dynamics, their performance is highly sensitive to partition granularity and task complexity. The results expose limitations in current LLMs capacity for representation alignment, highlighting the need for further research on robust alignment between language and internal agent representations.