# Human-like Category Learning by Injecting Ecological Priors from Large Language Models into Neural Networks

arXiv: https://arxiv.org/abs/2402.01821 · HTML: https://mlsanity.com/paper/2402.01821/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 76 references that are in the map.
- First posted on arXiv 2 Feb 2024.

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second](https://mlsanity.com/paper/2207.01848/) (2022)
- [Textbooks Are All You Need](https://mlsanity.com/paper/2306.11644/) (2023)
- [Learning to reinforcement learn](https://mlsanity.com/paper/1611.05763/) (2016)
- [Transformers Can Do Bayesian Inference](https://mlsanity.com/paper/2112.10510/) (2021)
- [Data Distributional Properties Drive Emergent In-Context Learning in Transformers](https://mlsanity.com/paper/2205.05055/) (2022)
- [Improving Text Embeddings with Large Language Models](https://mlsanity.com/paper/2401.00368/) (2023)
- [Language Models are Realistic Tabular Data Generators](https://mlsanity.com/paper/2210.06280/) (2022)
- [When Do Neural Nets Outperform Boosted Trees on Tabular Data?](https://mlsanity.com/paper/2305.02997/) (2023)
- [Generating Datasets with Pretrained Language Models](https://mlsanity.com/paper/2104.07540/) (2021)
- [OpenML Benchmarking Suites](https://mlsanity.com/paper/1708.03731/) (2017)
- [Orca 2: Teaching Small Language Models How to Reason](https://mlsanity.com/paper/2311.11045/) (2023)
- [Evaluating and Mitigating Discrimination in Language Model Decisions](https://mlsanity.com/paper/2312.03689/) (2023)
- [Meta-learning of Sequential Strategies](https://mlsanity.com/paper/1905.03030/) (2019)
- [Meta-in-context learning in large language models](https://mlsanity.com/paper/2305.12907/) (2023)
- [Cognitive Model Priors for Predicting Human Decisions](https://mlsanity.com/paper/1905.09397/) (2019)
- [Bayes in the age of intelligent machines](https://mlsanity.com/paper/2311.10206/) (2023)
- [Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines](https://mlsanity.com/paper/2205.11558/) (2022)
- [Modeling Human Exploration Through Resource-Rational Reinforcement Learning](https://mlsanity.com/paper/2201.11817/) (2022)
- [Cognition and Reality](https://mlsanity.com/paper/1802.09627/) (2017)
- [Building artificial neural circuits for domain-general cognition: a primer on brain-inspired systems-level architecture](https://mlsanity.com/paper/2303.13651/) (2023)

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

- [Generating Computational Cognitive Models using Large Language Models](https://mlsanity.com/paper/2502.00879/) (2025)
- [A Review of Incorporating Psychological Theories in LLMs](https://mlsanity.com/paper/2505.00003/) (2025)
- [Meta-learning ecological priors from large language models explains human learning and decision making](https://mlsanity.com/paper/2509.00116/) (2025)
- [Meta-learning as a principle for human-like visual representations](https://mlsanity.com/paper/2606.28399/) (2026)
- [More accurate behavioral predictions with hybrid Bayesian-connectionist models](https://mlsanity.com/paper/2608.22154/) (2026)

## Abstract (arXiv)

Ecological rationality refers to the notion that humans are rational agents adapted to their environment. However, testing this theory remains challenging due to two reasons: the difficulty in defining what tasks are ecologically valid and building rational models for these tasks. In this work, we demonstrate that large language models can generate cognitive tasks, specifically category learning tasks, that match the statistics of real-world tasks, thereby addressing the first challenge. We tackle the second challenge by deriving rational agents adapted to these tasks using the framework of meta-learning, leading to a class of models called ecologically rational meta-learned inference (ERMI). ERMI quantitatively explains human data better than seven other cognitive models in two different experiments. It additionally matches human behavior on a qualitative level: (1) it finds the same tasks difficult that humans find difficult, (2) it becomes more reliant on an exemplar-based strategy for assigning categories with learning, and (3) it generalizes to unseen stimuli in a human-like way. Furthermore, we show that ERMI's ecologically valid priors allow it to achieve state-of-the-art performance on the OpenML-CC18 classification benchmark.