# Towards Concept-Aware Large Language Models

arXiv: https://arxiv.org/abs/2311.01866 · HTML: https://mlsanity.com/paper/2311.01866/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 36 references that are in the map.
- First posted on arXiv 3 Nov 2023.

## Builds on (most cited first, 18 of 18)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [LaMDA: Language Models for Dialog Applications](https://mlsanity.com/paper/2201.08239/) (2022)
- [ConceptNet 5.5: An Open Multilingual Graph of General Knowledge](https://mlsanity.com/paper/1612.03975/) (2016)
- [Atlas: Few-shot Learning with Retrieval Augmented Language Models](https://mlsanity.com/paper/2208.03299/) (2022)
- [Surface Form Competition: Why the Highest Probability Answer Isn't Always Right](https://mlsanity.com/paper/2104.08315/) (2021)
- [Unsupervised Domain Clusters in Pretrained Language Models](https://mlsanity.com/paper/2004.02105/) (2020)
- [SenseBERT: Driving Some Sense into BERT](https://mlsanity.com/paper/1908.05646/) (2019)
- [HTLM: Hyper-Text Pre-Training and Prompting of Language Models](https://mlsanity.com/paper/2107.06955/) (2021)
- [ProtoQA: A Question Answering Dataset for Prototypical Common-Sense Reasoning](https://mlsanity.com/paper/2005.00771/) (2020)
- [Analyzing Encoded Concepts in Transformer Language Models](https://mlsanity.com/paper/2206.13289/) (2022)
- [Commonsense Properties from Query Logs and Question Answering Forums](https://mlsanity.com/paper/1905.10989/) (2019)
- [How Large Are Lions? Inducing Distributions over Quantitative Attributes](https://mlsanity.com/paper/1906.01327/) (2019)
- [Do language models have coherent mental models of everyday things?](https://mlsanity.com/paper/2212.10029/) (2022)

## Built on by (most cited first, 6 of 6)

- [From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning](https://mlsanity.com/paper/2505.17117/) (2025)
- [LLM-assisted Concept Discovery: Automatically Identifying and Explaining Neuron Functions](https://mlsanity.com/paper/2406.08572/) (2024)
- [The Philosophical Foundations of Growing AI Like A Child](https://mlsanity.com/paper/2502.10742/) (2025)
- [Vision-and-Language Training Helps Deploy Taxonomic Knowledge but Does Not Fundamentally Alter It](https://mlsanity.com/paper/2507.13328/) (2025)
- [What do MLLMs hear? Examining reasoning with text and sound components in Multimodal Large Language Models](https://mlsanity.com/paper/2406.04615/) (2024)
- [LaCy: What Small Language Models Can and Should Learn is Not Just a Question of Loss](https://mlsanity.com/paper/2602.12005/) (2026)

## Abstract (arXiv)

Concepts play a pivotal role in various human cognitive functions, including learning, reasoning and communication. However, there is very little work on endowing machines with the ability to form and reason with concepts. In particular, state-of-the-art large language models (LLMs) work at the level of tokens, not concepts. In this work, we analyze how well contemporary LLMs capture human concepts and their structure. We then discuss ways to develop concept-aware LLMs, taking place at different stages of the pipeline. We sketch a method for pretraining LLMs using concepts, and also explore the simpler approach that uses the output of existing LLMs. Despite its simplicity, our proof-of-concept is shown to better match human intuition, as well as improve the robustness of predictions. These preliminary results underscore the promise of concept-aware LLMs.