# Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines

arXiv: https://arxiv.org/abs/2205.11558 · HTML: https://mlsanity.com/paper/2205.11558/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 63 references that are in the map.
- First posted on arXiv 23 May 2022.

## Builds on (most cited first, 30 of 34)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Relational inductive biases, deep learning, and graph networks](https://mlsanity.com/paper/1806.01261/) (2018)
- [Meta-Learning in Neural Networks: A Survey](https://mlsanity.com/paper/2004.05439/) (2020)
- [Multimodal Few-Shot Learning with Frozen Language Models](https://mlsanity.com/paper/2106.13884/) (2021)
- [Reinforcement Learning with Unsupervised Auxiliary Tasks](https://mlsanity.com/paper/1611.05397/) (2016)
- [Learning to reinforcement learn](https://mlsanity.com/paper/1611.05763/) (2016)
- [The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision](https://mlsanity.com/paper/1904.12584/) (2019)
- [Inductive Biases for Deep Learning of Higher-Level Cognition](https://mlsanity.com/paper/2011.15091/) (2020)
- [Machine Theory of Mind](https://mlsanity.com/paper/1802.07740/) (2018)
- [Learning to Navigate in Complex Environments](https://mlsanity.com/paper/1611.03673/) (2016)
- [A Survey of Reinforcement Learning Informed by Natural Language](https://mlsanity.com/paper/1906.03926/) (2019)
- [Grounded Language Learning in a Simulated 3D World](https://mlsanity.com/paper/1706.06551/) (2017)
- [Grounding Referring Expressions in Images by Variational Context](https://mlsanity.com/paper/1712.01892/) (2017)
- [Learning with Latent Language](https://mlsanity.com/paper/1711.00482/) (2017)
- [Learning to Infer Graphics Programs from Hand-Drawn Images](https://mlsanity.com/paper/1707.09627/) (2017)
- [Learning Compositional Rules via Neural Program Synthesis](https://mlsanity.com/paper/2003.05562/) (2020)
- [Loss is its own Reward: Self-Supervision for Reinforcement Learning](https://mlsanity.com/paper/1612.07307/) (2016)
- [Meta-learning of Sequential Strategies](https://mlsanity.com/paper/1905.03030/) (2019)
- [Causal Reasoning from Meta-reinforcement Learning](https://mlsanity.com/paper/1901.08162/) (2019)
- [Environmental drivers of systematicity and generalization in a situated agent](https://mlsanity.com/paper/1910.00571/) (2019)
- [Leveraging Language to Learn Program Abstractions and Search Heuristics](https://mlsanity.com/paper/2106.11053/) (2021)
- [Tell me why! Explanations support learning relational and causal structure](https://mlsanity.com/paper/2112.03753/) (2021)
- [Universal linguistic inductive biases via meta-learning](https://mlsanity.com/paper/2006.16324/) (2020)
- [Gibbs Sampling with People](https://mlsanity.com/paper/2008.02595/) (2020)
- [Shaping Visual Representations with Language for Few-shot Classification](https://mlsanity.com/paper/1911.02683/) (2019)
- [Learning abstract structure for drawing by efficient motor program induction](https://mlsanity.com/paper/2008.03519/) (2020)

## Built on by (most cited first, 20 of 20)

- [Getting aligned on representational alignment](https://mlsanity.com/paper/2310.13018/) (2023)
- [Scaling Instructable Agents Across Many Simulated Worlds](https://mlsanity.com/paper/2404.10179/) (2024)
- [Language Instructed Reinforcement Learning for Human-AI Coordination](https://mlsanity.com/paper/2304.07297/) (2023)
- [The SocialAI School: Insights from Developmental Psychology Towards Artificial Socio-Cultural Agents](https://mlsanity.com/paper/2307.07871/) (2023)
- [Doing Experiments and Revising Rules with Natural Language and Probabilistic Reasoning](https://mlsanity.com/paper/2402.06025/) (2024)
- [Human-like Few-Shot Learning via Bayesian Reasoning over Natural Language](https://mlsanity.com/paper/2306.02797/) (2023)
- [Structure in Deep Reinforcement Learning: A Survey and Open Problems](https://mlsanity.com/paper/2306.16021/) (2023)
- [Reinforcement Learning with Simple Sequence Priors](https://mlsanity.com/paper/2305.17109/) (2023)
- [Seeing Like an AI: How LLMs Apply (and Misapply) Wikipedia Neutrality Norms](https://mlsanity.com/paper/2407.04183/) (2024)
- [Predictive auxiliary objectives in deep RL mimic learning in the brain](https://mlsanity.com/paper/2310.06089/) (2023)
- [Instilling Inductive Biases with Subnetworks](https://mlsanity.com/paper/2310.10899/) (2023)
- [Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networks](https://mlsanity.com/paper/2402.01821/) (2024)
- [Hierarchical reinforcement learning with natural language subgoals](https://mlsanity.com/paper/2309.11564/) (2023)
- [Comparing Abstraction in Humans and Large Language Models Using Multimodal Serial Reproduction](https://mlsanity.com/paper/2402.03618/) (2024)
- [Cognitive maps are generative programs](https://mlsanity.com/paper/2504.20628/) (2025)
- [Leveraging Systematic Knowledge of 2D Transformations](https://mlsanity.com/paper/2206.00893/) (2022)
- [Do as I can, not as I get](https://mlsanity.com/paper/2306.10345/) (2023)
- [Metareasoning constraints couple narratives, affect and cognition](https://mlsanity.com/paper/2502.09487/) (2025)
- [FairBelief -- Assessing Harmful Beliefs in Language Models](https://mlsanity.com/paper/2402.17389/) (2024)
- [Prospective Compression in Human Abstraction Learning](https://mlsanity.com/paper/2605.09985/) (2026)

## Abstract (arXiv)

Strong inductive biases give humans the ability to quickly learn to perform a variety of tasks. Although meta-learning is a method to endow neural networks with useful inductive biases, agents trained by meta-learning may sometimes acquire very different strategies from humans. We show that co-training these agents on predicting representations from natural language task descriptions and programs induced to generate such tasks guides them toward more human-like inductive biases. Human-generated language descriptions and program induction models that add new learned primitives both contain abstract concepts that can compress description length. Co-training on these representations result in more human-like behavior in downstream meta-reinforcement learning agents than less abstract controls (synthetic language descriptions, program induction without learned primitives), suggesting that the abstraction supported by these representations is key.