Dataset · Hugging Face

Sayanjib/numfmt-core

Name: numfmt-core
Creator: Sayanjib
License: cc-by-4.0
Keywords: task_categories:question-answering, task_categories:multiple-choice, language:en, license:cc-by-4.0, size_categories:1K<n<10K, region:us, number-formatting, format-sensitivity, small-language-models, benchmark, mcq

NumFmt-Core NumFmt-Core is a benchmark for measuring number-format sensitivity in language models. It contains 7 datasets rewritten into 5 mathematically equivalent number formats, used in the paper NumFmt-SLM: Does Number Formatting Affect Small Language Model Reasoning? (2025). Dataset structure Each file is named data/{dataset}__{format}.jsonl and contains one JSON object per line. Datasets Dataset Task N examples Split AQuA-RAT MCQ… See the full description on the dataset page: https://huggingface.co/datasets/Sayanjib/numfmt-core.

View on Hugging FaceReport an issue

Author / organization	Sayanjib
License (as declared)	cc-by-4.0
Downloads	142 (as reported by Hugging Face)
Likes	0
Created	29 May 2026
Last updated at source	29 May 2026

number-formattingformat-sensitivitysmall-language-modelsbenchmarkmcq

Embed this entry

<iframe src="https://silytics.si/embed/item/97c17630-d0a0-49aa-8e8d-a9596de134cc" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.