Xiezhi

Xiezhi is a Chinese-and-English multiple-choice benchmark covering holistic knowledge across 516 academic disciplines in 13 subjects.

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.

Category	knowledge
Subcategory	multi-disciplinary domain knowledge (multiple-choice)
Page status	active
Metric	accuracy
Direction	higher_is_better
Unit	percent
Dataset size	249587
Dataset licence	CC BY-NC-SA 4.0 (dataset); MIT (code)
Publisher	Fudan University and collaborators

What it measures

Xiezhi tests how much domain knowledge a model holds across a very wide span of academic and professional fields, in both Chinese and English, using multiple-choice questions adapted from exam and textbook material.

Task format

Multiple-choice question answering; each item gives a discipline-specific question and a candidate label pool, and the model selects the correct label.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub