The Digital Library of Mathematical Functions (DLMF) is an online reference on the properties of the special functions of applied mathematics. It contains three methodological chapters and 33 chapters focused on individual classes of functions. Information on functions include definitions, notations, series expansions, integrals, connection formulae, asymptotics, special values, methods of computation, etc. It includes hundreds of visualizations of the functions, including interactive three-dimensional representations of complex-valued functions. A math-aware search engine is also available. Extensive links to available software and external references are also provided.

INTELLECT-MATH: Frontier Mathematical Reasoning through Better Initializations for Reinforcement Learning

INTELLECT-MATH is a 7B parameter model optimized for mathematical reasoning. It was trained in two stages, an SFT stage, in which the model was fine-tuned on verified QwQ outputs, and an RL stage, in which the model was trained using the PRIME-RL recipe. We demonstrate that the quality of our SFT data can impact the performance and training speed of the RL stage: Due to its… See the full description on the dataset page: https://huggingface.co/datasets/PrimeIntellect/INTELLECT-MATH-SFT-Data.

Using the simple example of calculating the probability of reaching a traffic light while green, students are shown how to build a mathematical model using a very commonly-taught formula (sum of first n integers) to solve a rather practical problem. This resource is from PUMAS - Practical Uses of Math and Science - a collection of brief examples created by scientists and engineers showing how math and science topics taught in K-12 classes have real world applications.

MATH is a new dataset of 12,500 challenging competition mathematics problems. Each problem in MATH has a full step-by-step solution which can be used to teach models to generate answer derivations and explanations.

dev-store/math-diff-data dataset hosted on Hugging Face and contributed by the HF Datasets community

This large, international dataset contains survey responses from N = 12,570 students from 100 universities in 35 countries, collected in 21 languages. We measured anxieties (statistics, mathematics, test, trait, social interaction, performance, creativity, intolerance of uncertainty, and fear of negative evaluation), self-efficacy, persistence, and the cognitive reflection test, and collected demographics, previous mathematics grades, self-reported and official statistics grades, and statistics module details. Data reuse potential is broad, including testing links between anxieties and statistics/mathematics education factors, and examining instruments’ psychometric properties across different languages and contexts. Note that the pre-registration can be found here: https://osf.io/xs5wf

Materials associated with the manuscript "Do mathematicians and undergraduates agree about explanation quality?".

Raw judgement data from both groups are in the tab "RawJudgements-All" in the excel sheet. Each line represents a judgement by the judge listed in the Judge column. The candidates are identified by file names from the zip folder (which contains 10 pdf files, one for each of the 10 explanations).

Data were analysed using the AnalysisScriptForSharing.R script, separately for the mathematician and undergraduate groups.

This resulted in the results in the ModelledResults tab of the excel sheet. Column headings are: - individual: the explanation file name - math.Ntot: total number of judgements made by mathematicians for this explanation - math.N1: number of 'wins' by this explanation - math.N0: number of 'losses' by this explanation - math.theta: the beta parameter for this explanation derived from mathematicians' judgements - math.se.theta: the SE around the beta parameter derived from mathematicians' judgements - ug columns are all analogous to the math columns.

Abstract Issues involving the identification, analysis, and interpretation of mistakes made by mathematics students are not recent, although much still can be investigated on this subject. The aim of the present study was to identify, from the existing literature, relevant variables in the production of errors in mathematics. A systematic review of the literature of the period between 2012 and 2017 was independently performed by two researchers to evaluate the concordance between them. We searched the ERIC, PsycArticles, SciELO and Math Educ Database databases with the descriptors error AND mathematics OR math, error AND procedure AND mathematics OR math, error pattern AND mathematics OR math, analysis of errors AND mathematics OR math, systemic error AND mathematics OR math and their correspondents in Portuguese and Spanish. A total of 415 articles were identified, of which 31 were analyzed, dealing with error production. The variables identified as responsible for producing the most common errors refer to the student's internal causes or unspecified difficulties and errors in the teaching procedures. Responsibility for error is usually attributed to the students and the main trend of the research is only to inform the production of errors, since only a few studies have indicated ways to avoid or deal with errors produced by students in a specific and descriptive way. We emphasize importance and necessity of investigating educational practices to prevent and deal with errors.

the data for music effect on math performance

The 2017-18 tables are based on data collected from all of the nation’s school districts and schools—approximately 17,300 school districts and 96,300 schools. This set of Excel files contains data for schools offering courses and student enrollment data in mathematics and science courses for all states.

Data-Agora/math-seed-data dataset hosted on Hugging Face and contributed by the HF Datasets community

Data file for project on environmental impact and health

People from many countries have expressed interest in the tests students take for the Programme for International Student Assessment (PISA). Learning Mathematics for Life examines the link between the PISA test requirements and student performance. It focuses specifically on the proportions of students who answer questions correctly across a range of difficulty. The questions are classified by content, competencies, context and format, and the connections between these and student performance are then analysed. This analysis has been carried out in an effort to link PISA results to curricular programmes and structures in participating countries and economies. Results from the student assessment reflect differences in country performance in terms of the test questions. These findings are important for curriculum planners, policy makers and in particular teachers – especially mathematics teachers of intermediate and lower secondary school classes.

The data include information about 85 rigorous experimental studies that evaluated 64 programs in grades K-5 mathematics. These data were collected by the research team from studies included in a systematic review of programs for elementary mathematics. The data contain study and finding level information to examine what types of programs are most effective.

Abstract: This study analyzes the responses given by 332 8th grade Portuguese students to two questions, one with data set defined by a frequency table and the other defined by a bar graph. The goal is the evaluation of the effect of the data representation on the calculation of the median. Using a combined methodology, a frequency analysis of students' response types and a semiotic analysis of responses, based on onto-semiotic approach of knowledge and mathematics education, were applied. The semiotic analysis' main goal was the identification of objects and mathematical processes, which can characterize semiotic conflicts implied from the student responses. Generally, students revealed greater tendency to not respond to the graphic determination of the median, however, those who do answer, tend to obtain the median with less difficulty in a graphic context. Semiotic conflicts seem not to depend on the presentation of the data set, although there are more answers that are incorrect in the tabular context than in the graphic context.

Guatemala: PISA math scores: The latest value from 2022 is 344.199 index points, unavailable from index points in . In comparison, the world average is 439.569 index points, based on data from 78 countries. Historically, the average for Guatemala from 2022 to 2022 is 344.199 index points. The minimum value, 344.199 index points, was reached in 2022 while the maximum of 344.199 index points was recorded in 2022.

Abstract This article presents reflections on the viability of use of investigative tasks in mathematics teaching in elementary school through comprehension of strengths and weaknesses in this area of application, starting from a teacher's actions and hindsight. The data were obtained from the teacher and students (8 and 9-year-olds) of a forth-grade class from a Brazilian public elementary school. The main components of the data are the following: interviews, audio, and video recordings, a questionnaire and reports done by the teacher and students' records. Some of the identified potentialities are: encouragement to autonomy and creativity; development of students' argumentation and registering skills; search for different strategies mobilizing math knowledge. Limitations were also identified such as time to develop the tasks; break time between classes; children's lack of experience with argumentation and registering; unpredictability generated by the openness of the proposal. The potentialities raised point out to a possibility of using these tasks in the early grades, contributing not only to mathematics learning, but also to the development of competences such as respect and autonomy. The limitations observed were either overcome or minimized in many moments through the teacher's reflexive attitude and through team work.

Dataset Card

This dataset contains 23K conversations, serving as the official training data for the GenPRM model. The data synthesis and filtering procedures follow the methodology described in our paper:

Project Page: GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning Paper: https://arxiv.org/abs/2504.00891 Code: https://github.com/RyanLiu112/GenPRM

  Models

SFT models are available in the following repositories:

GenPRM-7B… See the full description on the dataset page: https://huggingface.co/datasets/GenPRM/GenPRM-MATH-Data.

Comprehensive dataset of 17 Mathematics schools in Washington, United States as of July, 2025. Includes verified contact information (email, phone), geocoded addresses, customer ratings, reviews, business categories, and operational details. Perfect for market research, lead generation, competitive analysis, and business intelligence. Download a complimentary sample to evaluate data quality and completeness.