# What Is Worth Representing? Representational Empowerment for Continual Model Construction

arXiv: https://arxiv.org/abs/2609.02322 · HTML: https://mlsanity.com/paper/2609.02322/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 64 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 30 of 37)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Recurrent World Models Facilitate Policy Evolution](https://mlsanity.com/paper/1809.01999/) (2018)
- [Diversity is All You Need: Learning Skills without a Reward Function](https://mlsanity.com/paper/1802.06070/) (2018)
- [Overcoming catastrophic forgetting with hard attention to the task](https://mlsanity.com/paper/1801.01423/) (2018)
- [Variational Intrinsic Control](https://mlsanity.com/paper/1611.07507/) (2016)
- [BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning](https://mlsanity.com/paper/1810.08272/) (2018)
- [Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task Planning](https://mlsanity.com/paper/2305.14909/) (2023)
- [Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning](https://mlsanity.com/paper/1509.08731/) (2015)
- [Scalable methods for computing state similarity in deterministic Markov Decision Processes](https://mlsanity.com/paper/1911.09291/) (2019)
- [A Laplacian Framework for Option Discovery in Reinforcement Learning](https://mlsanity.com/paper/1703.00956/) (2017)
- [WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment](https://mlsanity.com/paper/2402.12275/) (2024)
- [Maximum Entropy Gain Exploration for Long Horizon Multi-goal Reinforcement Learning](https://mlsanity.com/paper/2007.02832/) (2020)
- [Motif: Intrinsic Motivation from Artificial Intelligence Feedback](https://mlsanity.com/paper/2310.00166/) (2023)
- [The Value Equivalence Principle for Model-Based Reinforcement Learning](https://mlsanity.com/paper/2011.03506/) (2020)
- [Foundation Policies with Hilbert Representations](https://mlsanity.com/paper/2402.15567/) (2024)
- [Transfer in Deep Reinforcement Learning Using Successor Features and Generalised Policy Improvement](https://mlsanity.com/paper/1901.10964/) (2019)
- [Autotelic Agents with Intrinsically Motivated Goal-Conditioned Reinforcement Learning: a Short Survey](https://mlsanity.com/paper/2012.09830/) (2020)
- [Empowerment -- an Introduction](https://mlsanity.com/paper/1310.1863/) (2013)
- [Near Optimal Behavior via Approximate State Abstraction](https://mlsanity.com/paper/1701.04113/) (2017)
- [Learning Neuro-Symbolic Skills for Bilevel Planning](https://mlsanity.com/paper/2206.10680/) (2022)
- [Systematic Evaluation of Causal Discovery in Visual Model Based Reinforcement Learning](https://mlsanity.com/paper/2107.00848/) (2021)
- [Predicate Invention for Bilevel Planning](https://mlsanity.com/paper/2203.09634/) (2022)
- [People construct simplified mental representations to plan](https://mlsanity.com/paper/2105.06948/) (2021)
- [Human-Level Reinforcement Learning through Theory-Based Modeling, Exploration, and Planning](https://mlsanity.com/paper/2107.12544/) (2021)
- [Bisimulation Makes Analogies in Goal-Conditioned Reinforcement Learning](https://mlsanity.com/paper/2204.13060/) (2022)
- [Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models](https://mlsanity.com/paper/2507.12547/) (2025)

## Built on by (most cited first, 1 of 1)

- [Scaling Empowered Agents](https://mlsanity.com/paper/2610.13002/) (2026)

## Abstract (arXiv)

The first problem of modeling the world is not just estimating the right parameters or causal structure, but deciding what should be represented at all. We frame this problem as continual model construction: an agent maintains an environment-specific model M of an inaccessible world W and curates a persistent library L of reusable representational elements across environments. We propose Representational Empowerment (RepEmp) to score candidate elements by how much they expand the agent's future capacity to model and plan, complementing the classic definition of empowerment, but redefined as control over internal representations instead of external states. We realize the framework as a hierarchical Curator-Actor architecture and test it across three experiments. In a closed-vocabulary causal-learning task, human participants construct causal models at varying abstraction granularities to maximize goal reachability rather than fidelity to the world, a signature better predicted by RepEmp than by information-gain alternatives. Matched simulations reveal that RepEmp-guided construction contributes more than exploration to sufficient structure recovery and cross-task transfer. Finally, in an open-vocabulary planning domain, an LLM-augmented Curator builds more compact symbolic libraries, which also generalize better than baselines. Ablating RepEmp eliminates these benefits. Together, these results identify RepEmp as a key principle for continual model construction: deciding what to build, retain, and reuse under bounded resources.