Deriving Neural Scaling Laws from the statistics of natural language

• URL: http://arxiv.org/abs/2602.07488v2
• Date: Thu, 12 Feb 2026 11:54:22 GMT
• Title: Deriving Neural Scaling Laws from the statistics of natural language
• Authors: Francesco Cagnetta, Allan Raventós, Surya Ganguli, Matthieu Wyart,
• Abstract summary: We provide the first theory in the case of data-limited scaling laws.<n>We isolate two key statistical properties of language that alone can predict neural scaling exponents.<n>Our theory exhibits a remarkable match with experimentally measured neural scaling laws obtained from training GPT-2 and LLaMA style models.
• Score: 23.701814586453654
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite the fact that experimental neural scaling laws have substantially guided empirical progress in large-scale machine learning, no existing theory can quantitatively predict the exponents of these important laws for any modern LLM trained on any natural language dataset. We provide the first such theory in the case of data-limited scaling laws. We isolate two key statistical properties of language that alone can predict neural scaling exponents: (i) the decay of pairwise token correlations with time separation between token pairs, and (ii) the decay of the next-token conditional entropy with the length of the conditioning context. We further derive a simple formula in terms of these statistics that predicts data-limited neural scaling exponents from first principles without any free parameters or synthetic data models. Our theory exhibits a remarkable match with experimentally measured neural scaling laws obtained from training GPT-2 and LLaMA style models from scratch on two qualitatively different benchmarks, TinyStories and WikiText.

Related papers

• On the origin of neural scaling laws: from random graphs to natural language [10.425020020850402]
  We study scaling laws for transformers trained to predict random walks (bigrams) on graphs with tunable complexity.<n>We consider dialing down the complexity of natural language systematically, by training on sequences sampled from increasingly simplified generative language models.<n>Our results also include scaling laws obtained from training on random walks on random graphs drawn from Erds-Renyi and scale-free Barabsi-Albert ensembles.
  arXiv  Detail & Related papers  (2026-01-15T18:46:09Z)
• I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data? [76.15163242945813]
  Large language models (LLMs) have led many to conclude that they exhibit a form of intelligence.<n>We introduce a novel generative model that generates tokens on the basis of human-interpretable concepts represented as latent discrete variables.
  arXiv  Detail & Related papers  (2025-03-12T01:21:17Z)
• Neural Scaling Laws Rooted in the Data Distribution [0.0]
  Deep neural networks exhibit empirical neural scaling laws, with error decreasing as a power law with increasing model or data size.<n>We develop a mathematical model intended to describe natural datasets using percolation theory.<n>We test the theory by training regression models on toy datasets derived from percolation theory simulations.
  arXiv  Detail & Related papers  (2024-12-10T22:01:38Z)
• Information-Theoretic Foundations for Neural Scaling Laws [20.617552198581024]
  We develop information-theoretic foundations for neural scaling laws. We observe that the optimal relation between data and model size is linear, up to logarithmic factors.
  arXiv  Detail & Related papers  (2024-06-28T02:20:54Z)
• A Dynamical Model of Neural Scaling Laws [79.59705237659547]
  We analyze a random feature model trained with gradient descent as a solvable model of network training and generalization. Our theory shows how the gap between training and test loss can gradually build up over time due to repeated reuse of data.
  arXiv  Detail & Related papers  (2024-02-02T01:41:38Z)
• The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute [66.84421705029624]
  We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours. We pre-process an existing large, diverse, and high-quality dataset of books that surpasses existing academic benchmarks in quality, diversity, and document length. This work also provides two baseline models: a feed-forward model derived from the GPT-2 architecture and a recurrent model in the form of a novel LSTM with ten-fold throughput.
  arXiv  Detail & Related papers  (2023-09-20T10:31:17Z)
• An Information-Theoretic Analysis of Compute-Optimal Neural Scaling Laws [24.356906682593532]
  We study the compute-optimal trade-off between model and training data set sizes for large neural networks. Our result suggests a linear relation similar to that supported by the empirical analysis of chinchilla.
  arXiv  Detail & Related papers  (2022-12-02T18:46:41Z)
• A Solvable Model of Neural Scaling Laws [72.8349503901712]
  Large language models with a huge number of parameters, when trained on near internet-sized number of tokens, have been empirically shown to obey neural scaling laws. We propose a statistical model -- a joint generative data model and random feature model -- that captures this neural scaling phenomenology. Key findings are the manner in which the power laws that occur in the statistics of natural datasets are extended by nonlinear random feature maps.
  arXiv  Detail & Related papers  (2022-10-30T15:13:18Z)
• The Neural Coding Framework for Learning Generative Models [91.0357317238509]
  We propose a novel neural generative model inspired by the theory of predictive processing in the brain. In a similar way, artificial neurons in our generative model predict what neighboring neurons will do, and adjust their parameters based on how well the predictions matched reality.
  arXiv  Detail & Related papers  (2020-12-07T01:20:38Z)
• Multi-timescale Representation Learning in LSTM Language Models [69.98840820213937]
  Language models must capture statistical dependencies between words at timescales ranging from very short to very long. We derived a theory for how the memory gating mechanism in long short-term memory language models can capture power law decay. Experiments showed that LSTM language models trained on natural English text learn to approximate this theoretical distribution.
  arXiv  Detail & Related papers  (2020-09-27T02:13:38Z)
• Parsimonious neural networks learn interpretable physical laws [77.34726150561087]
  We propose parsimonious neural networks (PNNs) that combine neural networks with evolutionary optimization to find models that balance accuracy with parsimony. The power and versatility of the approach is demonstrated by developing models for classical mechanics and to predict the melting temperature of materials from fundamental properties.
  arXiv  Detail & Related papers  (2020-05-08T16:15:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.