Learning is Forgetting: LLM Training As Lossy Compression

View PDF HTML (experimental) Abstract:Despite the increasing prevalence of large language models (LLMs), we still have a limited understanding of how their representational spaces are structured. This limits our ability to interpret how and what they learn or relate them to learning in humans. We argue LLMs are best seen as an instance of lossy compression, where over training they learn by retaining only information in their training data relevant to their objective(s). We show pre-training results in models that are optimally compressed for next-sequence prediction, approaching the Information Bottleneck bound on compression. Across an array of open weights models, each compresses differently, likely due to differences in the data and training recipes used. However even across different families of LLMs the optimality of a model's compression, and the information present in it, can predict downstream performance on across a wide array of benchmarks, letting us directly link representational structure to actionable insights about model performance. In the general case the work presented here offers a unified Information-Theoretic framing for how these models learn that is deployable at scale. Comments: 12 page core paper, 16 page Appendix - A shorter version with fewer visuals appears at ICLR 2026 Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Theory (cs.IT) Cite as: arXiv:2604.07569 [cs.LG] (or arXiv:2604.07569v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2604.07569 arXiv-issued DOI via DataCite (pending registration) Submission history From: Henry Conklin [view email] [v1] Wed, 8 Apr 2026 20:12:07 UTC (17,178 KB)