# Generalization through Lexical Abstraction in Transformer Models: The Case of Functional Words

arXiv: https://arxiv.org/abs/2609.19887 · HTML: https://mlsanity.com/paper/2609.19887/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 27 references that are in the map.
- First posted on arXiv 17 Sep 2026.

## Builds on (most cited first, 11 of 11)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [MPNet: Masked and Permuted Pre-training for Language Understanding](https://mlsanity.com/paper/2004.09297/) (2020)
- [State-of-the-art generalisation research in NLP: A taxonomy and review](https://mlsanity.com/paper/2210.03050/) (2022)
- [AbsPyramid: Benchmarking the Abstraction Ability of Language Models with a Unified Entailment Graph](https://mlsanity.com/paper/2311.09174/) (2023)
- [Structural Priming Demonstrates Abstract Grammatical Representations in Multilingual Language Models](https://mlsanity.com/paper/2311.09194/) (2023)
- [How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure](https://mlsanity.com/paper/2311.04900/) (2023)
- [Blackbird language matrices (BLM), a new task for rule-like generalization in neural networks: Motivations and Formal Specifications](https://mlsanity.com/paper/2306.11444/) (2023)
- [Are there identifiable structural parts in the sentence embedding whole?](https://mlsanity.com/paper/2406.16563/) (2024)
- [Investigating grammatical abstraction in language models using few-shot learning of novel noun gender](https://mlsanity.com/paper/2403.10338/) (2024)

## Abstract (arXiv)

Pronouns, adverbs and other functional words (such as they, her, somewhere, there) are often used in language to replace concrete nouns or phrases, when their properties - such as gender, grammatical number - provide sufficient information for the given context. Do pretrained transformer models encode such functional words in a manner that allows them to be used like humans do? Can language models recognize the syntactic and semantic parallelism of sentences such as "The researchers wrote the paper" and "They wrote it", which relies on such lexical abstraction? We map these linguistic questions into the embedding space of a pretrained transformer model, and compare representations of nouns, with the representations of the pronouns and adverbs that can replace these nouns, in isolation and in parallel lexicalized and functional sentences. We then probe for shared syntactic and semantic structure in the embeddings of parallel lexicalized and functional sentences. We find that functional words are located centrally compared to nouns, but are also distinct, which is congruent with their behaviour as place-holders in a wide variety of contexts. The analysis of the embeddings of parallel (lexicalized and functional) sentences show them inhabiting different subspaces of the embedding space. Experiments that distil the structural information of the sentence show that training on either type of data does not reveal the shared structure - because of the over-consistency of the vocabulary (in case of the functional data), and the too much variety (in case of the lexicalized versions). However, training with a mix of functional and lexicalized sentences, the shared structure emerges.