Nothing from Something: Can a Language Model Discover 0?

• URL: http://arxiv.org/abs/2606.17289v2
• Date: Thu, 18 Jun 2026 21:31:54 GMT
• Title: Nothing from Something: Can a Language Model Discover 0?
• Abstract summary: We use simple arithmetic as a case study for examining how modern AI models could expand their mathematical horizons.<n>We show that language models of a GPT-2 size are unable to perform this generalization at test time regardless of language pretraining.<n>We find that language pretraining reduces the number of required examples by approximately $50%$, showing that language abilities can scaffold mathematical discovery in neural models.
• Score: 9.459481698149597
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: AI systems based on artificial neural networks are being developed with aspirations of pushing the boundary of human mathematical knowledge. A key question for these systems is how much they can reach beyond their training data. Mathematical discovery requires a strong form of out of distribution generalization; the ability to hypothesize genuinely new - and potentially logically more powerful - mathematical structures. It has been hypothesized that language abilities support such generalizations in human cognition. In this work, we use simple arithmetic as a case study for examining how modern AI models could expand their mathematical horizons, evaluating whether these models can independently discover the concept of "zero". We show that (1) language models of a GPT-2 size are unable to perform this generalization at test time regardless of language pretraining, but (2) models can improve substantially after training on tens or hundreds of examples of zero. Additionally, we find that language pretraining reduces the number of required examples by approximately $50\%$, showing that language abilities can scaffold mathematical discovery in neural models.

Related papers

• Selecting Language Models for Social Science: Start Small, Start Open, and Validate [0.3823356975862005]
  We argue that social scientists cannot altogether avoid validating computational measures (ex-post)<n>Being able to reliably replicate a particular finding that entails the use of a language model requires reliably reproducing a task.<n>We propose starting with smaller, open models, and constructing benchmarks to demonstrate the validity of the entire computational pipeline.
  arXiv  Detail & Related papers  (2026-01-16T01:01:47Z)
• AI Agents as Universal Task Solvers [94.49762121230042]
  We show that the optimal speed-up that a universal solver can achieve using past data is tightly related to their algorithmic information.<n>We argue that the key quantity to optimize when scaling reasoning models is time, whose critical role in learning has so far only been indirectly considered.
  arXiv  Detail & Related papers  (2025-10-14T02:17:54Z)
• From Prediction to Understanding: Will AI Foundation Models Transform Brain Science? [37.27364085324663]
  Generative pretraining (the "GPT" in ChatGPT) enables language models to learn from vast amounts of internet text without human supervision.<n>We use the term foundation models to refer to large pretrained systems that can be adapted to a wide range of tasks within and across domains.<n>These models achieve strong predictive accuracy, raising hopes that they might illuminate computational principles.<n>Here, we outline how foundation models can be productively integrated into the brain sciences, highlighting both their promise and their limitations.
  arXiv  Detail & Related papers  (2025-09-21T23:39:04Z)
• Language Models Encode the Value of Numbers Linearly [28.88044346200171]
  We study how language models encode the value of numbers, a basic element in math. Experimental results support the existence of encoded number values in large language models. Our research provides evidence that LLMs encode the value of numbers linearly.
  arXiv  Detail & Related papers  (2024-01-08T08:54:22Z)
• Large Language Models for Scientific Synthesis, Inference and Explanation [56.41963802804953]
  We show how large language models can perform scientific synthesis, inference, and explanation. We show that the large language model can augment this "knowledge" by synthesizing from the scientific literature. This approach has the further advantage that the large language model can explain the machine learning system's predictions.
  arXiv  Detail & Related papers  (2023-10-12T02:17:59Z)
• Physics of Language Models: Part 3.2, Knowledge Manipulation [51.68385617116854]
  This paper investigates four fundamental knowledge manipulation tasks. We show that language models excel in knowledge retrieval but struggle even in the simplest classification or comparison tasks. Our findings also apply to modern pretrained language models such as GPT-4.
  arXiv  Detail & Related papers  (2023-09-25T17:50:41Z)
• Are Emergent Abilities in Large Language Models just In-Context Learning? [46.561464069450444]
  We present a novel theory that explains emergent abilities, taking into account their potential confounding factors. Our findings suggest that purported emergent abilities are not truly emergent, but result from a combination of in-context learning, model memory, and linguistic knowledge.
  arXiv  Detail & Related papers  (2023-09-04T20:54:11Z)
• A Sentence is Worth a Thousand Pictures: Can Large Language Models Understand Hum4n L4ngu4ge and the W0rld behind W0rds? [2.7342737448775534]
  Large Language Models (LLMs) have been linked to claims about human-like linguistic performance. We analyze the contribution of LLMs as theoretically informative representations of a target cognitive system. We evaluate the models' ability to see the bigger picture, through top-down feedback from higher levels of processing.
  arXiv  Detail & Related papers  (2023-07-26T18:58:53Z)
• State space models can express n-gram languages [51.823427608117626]
  We build state space language models that can solve the next-word prediction task for languages generated from n-gram rules.<n>Our proof shows how SSMs can encode n-gram rules using new theoretical results on their capacity.<n>We conduct experiments with a small dataset generated from n-gram rules to show how our framework can be applied to SSMs and RNNs obtained through gradient-based optimization.
  arXiv  Detail & Related papers  (2023-06-20T10:41:23Z)
• Modeling rapid language learning by distilling Bayesian priors into artificial neural networks [18.752638142258668]
  We show that learning from limited naturalistic data is possible with an approach that combines the strong inductive biases of a Bayesian model with the flexible representations of a neural network. The resulting system can learn formal linguistic patterns from a small number of examples. It can also learn aspects of English syntax from a corpus of natural language.
  arXiv  Detail & Related papers  (2023-05-24T04:11:59Z)
• Emergent Analogical Reasoning in Large Language Models [1.5469452301122177]
  We show that GPT-3 has a surprisingly strong capacity for abstract pattern induction, matching or even surpassing human capabilities in most settings. Our results indicate that large language models such as GPT-3 have acquired an emergent ability to find zero-shot solutions to a broad range of analogy problems.
  arXiv  Detail & Related papers  (2022-12-19T00:04:56Z)
• Discovering Latent Knowledge in Language Models Without Supervision [72.95136739040676]
  Existing techniques for training language models can be misaligned with the truth. We propose directly finding latent knowledge inside the internal activations of a language model in a purely unsupervised way. We show that despite using no supervision and no model outputs, our method can recover diverse knowledge represented in large language models.
  arXiv  Detail & Related papers  (2022-12-07T18:17:56Z)
• Towards Zero-shot Language Modeling [90.80124496312274]
  We construct a neural model that is inductively biased towards learning human languages. We infer this distribution from a sample of typologically diverse training languages. We harness additional language-specific side information as distant supervision for held-out languages.
  arXiv  Detail & Related papers  (2021-08-06T23:49:18Z)
• Leap-Of-Thought: Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge [96.92252296244233]
  Large pre-trained language models (LMs) acquire some reasoning capacity, but this ability is difficult to control. We show that LMs can be trained to reliably perform systematic reasoning combining both implicit, pre-trained knowledge and explicit natural language statements. Our work paves a path towards open-domain systems that constantly improve by interacting with users who can instantly correct a model by adding simple natural language statements.
  arXiv  Detail & Related papers  (2020-06-11T17:02:20Z)

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.