MathQA

A 37k-problem multiple-choice math word problem dataset built by re-annotating AQuA-RAT with interpretable, formal operation programs rather than free-text rationales.

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorymath
Subcategorymultiple-choice math word problems with operation-program annotations
Page statusactive
Metricaccuracy and length-normalised accuracy (acc, acc_norm) over the selected option
Directionhigher_is_better
Unit%
Dataset size37200
Dataset licenceApache-2.0
PublisherUniversity of Washington; Allen Institute for AI

What it measures

MathQA gives a model a short math word problem in English plus five lettered answer options and asks it to pick the correct one. The problems themselves come from the earlier AQuA-RAT dataset; MathQA's contribution is a new representation language that annotates each problem with a formal, step-by-step operation program (a sequence of arithmetic operations over the problem's numbers and a small set of constants) rather than AQuA-RAT's original free-text rationale, so that a system's reasoning can be checked structurally rather than only judged by its final answer. As distributed for language-model evaluation, only the question and options are used; the operation-program annotations exist in the data but the standard multiple-choice harness task does not require a model to produce one.

Task format

Multiple-choice question answering: a word problem followed by (typically) five lettered options. In lm-evaluation-harness the task is scored by comparing the model's log-likelihood on each option's text as a completion of the question, not by free-text generation.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub