# Knowledge Acquisition During Pre-training? Large Language Models Learn Better With Auxiliary Views

arXiv: https://arxiv.org/abs/2609.04180 · HTML: https://mlsanity.com/paper/2609.04180/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 48 references that are in the map.
- First posted on arXiv 3 Sep 2026.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [How Can We Know What Language Models Know?](https://mlsanity.com/paper/1911.12543/) (2019)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [DataComp-LM: In search of the next generation of training sets for language models](https://mlsanity.com/paper/2406.11794/) (2024)
- [Physics of Language Models: Part 3.1, Knowledge Storage and Extraction](https://mlsanity.com/paper/2309.14316/) (2023)
- [Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?](https://mlsanity.com/paper/2405.05904/) (2024)
- [A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity](https://mlsanity.com/paper/2305.13169/) (2023)
- [Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs](https://mlsanity.com/paper/2312.05934/) (2023)
- [Towards Continual Knowledge Learning of Language Models](https://mlsanity.com/paper/2110.03215/) (2021)
- [Infini-gram: Scaling Unbounded n-gram Language Models to a Trillion Tokens](https://mlsanity.com/paper/2401.17377/) (2024)
- [Scaling Laws and Interpretability of Learning from Repeated Data](https://mlsanity.com/paper/2205.10487/) (2022)
- [SaulLM-7B: A pioneering Large Language Model for Law](https://mlsanity.com/paper/2403.03883/) (2024)
- [To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis](https://mlsanity.com/paper/2305.13230/) (2023)
- [Reuse, Don't Retrain: A Recipe for Continued Pretraining of Language Models](https://mlsanity.com/paper/2407.07263/) (2024)
- [Can Generative Pre-trained Language Models Serve as Knowledge Bases for Closed-book QA?](https://mlsanity.com/paper/2106.01561/) (2021)
- [Instruction-tuned Language Models are Better Knowledge Learners](https://mlsanity.com/paper/2402.12847/) (2024)
- [Meta-Learning Online Adaptation of Language Models](https://mlsanity.com/paper/2305.15076/) (2023)
- [TxGemma: Efficient and Agentic LLMs for Therapeutics](https://mlsanity.com/paper/2504.06196/) (2025)
- [Towards Objective Fine-tuning: How LLMs' Prior Knowledge Causes Potential Poor Calibration?](https://mlsanity.com/paper/2505.20903/) (2025)

## Abstract (arXiv)

Gaps remain in our understanding of how large language models (LLMs) acquire knowledge during pre-training. We posit that auxiliary views, reformulations of knowledge, are causally helpful for learning. We design controlled experiments to isolate this. First, we confirm that repetition is necessary for acquisition and clarify that paraphrasing helps only at smaller batch sizes. Second, holding the token budget fixed, allocating tokens from document repetition to auxiliary views improves learning, counterintuitively, even for factual recall. Third, the effectiveness of auxiliary views is not contingent on the strength of the teacher model that generates them. Fourth, we identify forms of knowledge, contextual and foundational, that aid learning in the presence of prior knowledge gaps. Finally, we examine how these effects manifest mechanistically via layer-wise biases and compression. Together, our findings suggest that auxiliary representations of knowledge, which arise naturally in large pre-training corpora, are a key factor in the success of pre-training and offer a plausible explanation for why data diversity matters.