# Continual learning and refinement of causal models through dynamic predicate invention

arXiv: https://arxiv.org/abs/2602.17217 · HTML: https://mlsanity.com/paper/2602.17217/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World models for language agents, one of 11 directions in World models.
- Builds on 12 of its 23 references that are in the map.
- First posted on arXiv 19 Feb 2026.

## Placed in

- World models / [World models for language agents](https://mlsanity.com/wm/world-models-for-language-agents/)

## Builds on (most cited first, 12 of 12)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Learning programs by learning from failures](https://mlsanity.com/paper/2005.02259/) (2020)
- [Learning Symbolic Operators for Task and Motion Planning](https://mlsanity.com/paper/2103.00589/) (2021)
- [Learning Neuro-Symbolic Skills for Bilevel Planning](https://mlsanity.com/paper/2206.10680/) (2022)
- [Making sense of sensory input](https://mlsanity.com/paper/1910.02227/) (2019)
- [PoE-World: Compositional World Modeling with Products of Programmatic Experts](https://mlsanity.com/paper/2505.10819/) (2025)
- [Explainable Reinforcement Learning via a Causal World Model](https://mlsanity.com/paper/2305.02749/) (2023)
- [Logical reduction of metarules](https://mlsanity.com/paper/1907.10952/) (2019)
- [Inductive general game playing](https://mlsanity.com/paper/1906.09627/) (2019)
- [From Pixels to Predicates: Learning Symbolic World Models via Pretrained Vision-Language Models](https://mlsanity.com/paper/2501.00296/) (2024)
- [Top Program Construction and Reduction for polynomial time Meta-Interpretive Learning](https://mlsanity.com/paper/2101.05050/) (2021)
- [Relational Neurosymbolic Markov Models](https://mlsanity.com/paper/2412.13023/) (2024)

## Abstract (arXiv)

Efficiently navigating complex environments requires agents to internalize the underlying logic of their world, yet standard world modelling methods often struggle with sample inefficiency, lack of transparency, and poor scalability. We propose a framework for constructing symbolic causal world models entirely online by integrating continuous model learning and repair into the agent's decision loop, by leveraging the power of Meta-Interpretive Learning and predicate invention to find semantically meaningful and reusable abstractions, allowing an agent to construct a hierarchy of disentangled, high-quality concepts from its observations. We demonstrate that our lifted inference approach scales to domains with complex relational dynamics, where propositional methods suffer from combinatorial explosion, while achieving sample-efficiency orders of magnitude higher than the established PPO neural-network-based baseline.