# What do Large Language Models Learn beyond Language?

arXiv: https://arxiv.org/abs/2210.12302 · HTML: https://mlsanity.com/paper/2210.12302/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 32 references that are in the map.
- First posted on arXiv 21 Oct 2022.

## Builds on (most cited first, 25 of 25)

- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [Deep contextualized word representations](https://mlsanity.com/paper/1802.05365/) (2018)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems](https://mlsanity.com/paper/1905.00537/) (2019)
- [Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation](https://mlsanity.com/paper/1609.08144/) (2016)
- [CodeBERT: A Pre-Trained Model for Programming and Natural Languages](https://mlsanity.com/paper/2002.08155/) (2020)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering](https://mlsanity.com/paper/1602.01585/) (2016)
- [Do NLP Models Know Numbers? Probing Numeracy in Embeddings](https://mlsanity.com/paper/1909.07940/) (2019)
- [Probing Pretrained Language Models for Lexical Semantics](https://mlsanity.com/paper/2010.05731/) (2020)
- [Open Sesame: Getting Inside BERT's Linguistic Knowledge](https://mlsanity.com/paper/1906.01698/) (2019)
- [On the Practical Computational Power of Finite Precision RNNs for Language Recognition](https://mlsanity.com/paper/1805.04908/) (2018)
- [Perturbed Masking: Parameter-free Probing for Analyzing and Interpreting BERT](https://mlsanity.com/paper/2004.14786/) (2020)
- [Representing Numbers in NLP: a Survey and a Vision](https://mlsanity.com/paper/2103.13136/) (2021)
- [Does syntax need to grow on trees? Sources of hierarchical inductive bias in sequence-to-sequence networks](https://mlsanity.com/paper/2001.03632/) (2020)
- [Positional Encoding as Spatial Inductive Bias in GANs](https://mlsanity.com/paper/2012.05217/) (2020)
- [Superbizarre Is Not Superb: Derivational Morphology Improves BERT's Interpretation of Complex Words](https://mlsanity.com/paper/2101.00403/) (2021)
- [Investigating BERT's Knowledge of Language: Five Analysis Methods with NPIs](https://mlsanity.com/paper/1909.02597/) (2019)
- [LSTM Networks Can Perform Dynamic Counting](https://mlsanity.com/paper/1906.03648/) (2019)
- [Multi-SimLex: A Large-Scale Evaluation of Multilingual and Cross-Lingual Lexical Semantic Similarity](https://mlsanity.com/paper/2003.04866/) (2020)
- [What they do when in doubt: a study of inductive biases in seq2seq learners](https://mlsanity.com/paper/2006.14953/) (2020)
- [On the Inductive Bias of Dropout](https://mlsanity.com/paper/1412.4736/) (2014)
- [Quantifying the Contextualization of Word Representations with Semantic Class Probing](https://mlsanity.com/paper/2004.12198/) (2020)
- [On Evaluating the Generalization of LSTM Models in Formal Languages](https://mlsanity.com/paper/1811.01001/) (2018)
- [Evaluating the Ability of LSTMs to Learn Context-Free Grammars](https://mlsanity.com/paper/1811.02611/) (2018)

## Built on by (most cited first, 3 of 3)

- [Do Large Language Models Have Compositional Ability? An Investigation into Limitations and Scalability](https://mlsanity.com/paper/2407.15720/) (2024)
- [ICSVR: Investigating Compositional and Syntactic Understanding in Video Retrieval Models](https://mlsanity.com/paper/2306.16533/) (2023)
- [Understanding the Role of Input Token Characters in Language Models: How Does Information Loss Affect Performance?](https://mlsanity.com/paper/2310.17271/) (2023)

## Abstract (arXiv)

Large language models (LMs) have rapidly become a mainstay in Natural Language Processing. These models are known to acquire rich linguistic knowledge from training on large amounts of text. In this paper, we investigate if pre-training on text also confers these models with helpful `inductive biases' for non-linguistic reasoning. On a set of 19 diverse non-linguistic tasks involving quantitative computations, recognizing regular expressions and reasoning over strings. We find that pretrained models significantly outperform comparable non-pretrained neural models. This remains true also in experiments with training non-pretrained models with fewer parameters to account for model regularization effects. We further explore the effect of text domain on LMs by pretraining models from text from different domains and provenances. Our experiments surprisingly reveal that the positive effects of pre-training persist even when pretraining on multi-lingual text or computer code, and even for text generated from synthetic languages. Our findings suggest a hitherto unexplored deep connection between pre-training and inductive learning abilities of language models.