mlsanityData to 7 Oct 2026

Numeracy for Language Models: Evaluating and Improving their Ability to Predict Numbers

arXiv 1805.08154 ↗First posted 21 May 2018cs.CL cs.NE stat.MLcited by 42

Not on the map. Why?

Abstract (arXiv)

Numeracy is the ability to understand and work with numbers. It is a necessary skill for composing and understanding documents in clinical, scientific, and other technical domains. In this paper, we explore different strategies for modelling numerals with language models, such as memorisation and digit-by-digit composition, and propose a novel neural architecture that uses a continuous probability density function to model numerals from an open vocabulary. Our evaluation on clinical and scientific datasets shows that using hierarchical models to distinguish numerals from words improves a perplexity metric on the subset of numerals by 2 and 4 orders of magnitude, respectively, over non-hierarchical models. A combination of strategies can further improve perplexity. Our continuous probability density function model reduces mean absolute percentage errors by 18% and 54% in comparison to the second best strategy for each dataset, respectively.

Key facts. Cited by 42 papers in the map (data through 7 Oct 2026). First posted on arXiv 21 May 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 9; 2023: 4; 2024: 13; 2025: 5; 2026: 7.

42 papers.

Sort:

By area:

  1. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 518
  2. 2022Impact of Pretraining Term Frequencies on Few-Shot Reasoningcited by 142
  3. 2024Why Tabular Foundation Models Should Be a Research Prioritycited by 103
  4. 2021Representing Numbers in NLP: a Survey and a Visioncited by 83
  5. 2022Probing for the Usage of Grammatical Numbercited by 57
  6. 2022Leveraging Language Foundation Models for Human Mobility Forecastingcited by 50
  7. 2024Number Cookbook: Number Understanding of Language Models and How to Improve Itcited by 41
  8. 2021Zero-shot Generalization in Dialog State Tracking through Generative Question Answeringcited by 33
  9. 2024Can Language Beat Numerical Regression? Language-Based Multimodal Trajectory Predictioncited by 31
  10. 2023Mathematical Language Models: A Surveycited by 24
  11. 2021How Much Coffee Was Consumed During EMNLP 2019? Fermi Problems: A New Reasoning Challenge for AIcited by 22
  12. 2024AutoElicit: Using Large Language Models for Expert Prior Elicitation in Predictive Modellingcited by 15
30 more
  1. 2025A Survey on Mathematical Reasoning and Optimization with Large Language Modelscited by 12
  2. 2023Human Behavioral Benchmarking: Numeric Magnitude Comparison Effects in Large Language Modelscited by 10
  3. 2024The Geometry of Numerical Reasoning: Language Models Compare Numeric Properties in Linear Subspacescited by 10
  4. 2022LUNA: Language Understanding with Number Augmentations on Transformers via Number Plugins and Pre-trainingcited by 8
  5. 2022Towards Robust Numerical Question Answering: Diagnosing Numerical Capabilities of NLP Systemscited by 6
  6. 2021Masked Measurement Prediction: Learning to Jointly Predict Quantities and Units from Textual Contextcited by 5
  7. 2023Learning Non-linguistic Skills without Sacrificing Linguistic Proficiencycited by 5
  8. 2024Knowledge of Pretrained Language Models on Surface Information of Tokenscited by 5
  9. 2024How to Leverage Digit Embeddings to Represent Numbers?cited by 5
  10. 2022Overcoming Barriers to Skill Injection in Language Modeling: Case Study in Arithmeticcited by 4
  11. 2025MachineLearningLM: Scaling Many-shot In-context Learning via Continued Pretrainingcited by 3
  12. 2022Do Language Models Understand Measurements?cited by 2
  13. 2023KitchenScale: Learning to predict ingredient quantities from recipe contextscited by 2
  14. 2024GPT Czech Poet: Generation of Czech Poetic Strophes with Language Modelscited by 2
  15. 2024Confidence-Aware Document OCR Error Detectioncited by 2
  16. 2024Do You Know What You Are Talking About? Characterizing Query-Knowledge Relevance For Reliable Retrieval Augmented Generationcited by 2
  17. 2022Improving Downstream Task Performance by Treating Numbers as Entitiescited by 1
  18. 2024Laying Anchors: Semantically Priming Numerals in Language Modelingcited by 1
  19. 2024Numbers Matter! Bringing Quantity-awareness to Retrieval Systemscited by 1
  20. 2025Agentic Physical AI toward a Domain-Specific Foundation Model for Energy Systems: A Case Study on Nuclear Reactor Controlcited by 1
  21. 2024Pre-Calc: Learning to Use the Calculator Improves Numeracy in Language Modelscited by 0
  22. 2025Disambiguating Numeral Sequences to Decipher Ancient Accounting Corporacited by 0
  23. 2025Wiki-Quantities and Wiki-Measurements: Datasets of Quantities and their Measurement Context from Wikipediacited by 0
  24. 2026Neural Isomorphic Fields: A Transformer-based Algebraic Numerical Embeddingcited by 0
  25. 2026CONE: Embeddings for Complex Numerical Data Preserving Unit and Variable Semanticscited by 0
  26. 2026PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Modelscited by 0
  27. 2026Physics-Guided Sequence-Based Generative Framework for Acoustic Metamaterial Inverse Designcited by 0
  28. 2026Toten: A Knowledge-Based System For Structure-Preserving Representation Of Physical Quantities And Technical Notation In Brazilian Portuguesecited by 0
  29. 2026Enhancing Numerical Prediction in LLMs via Smooth MMD Alignmentcited by 0
  30. 2026How Much is a Human Right Worth? ECtHR-NPD: A Benchmark for Predicting Non-Pecuniary Damage Awardscited by 0