FewCLUE: CSL (Keyword Recognition)

FewCLUE's keyword-recognition task: judge whether every keyword listed for a Chinese academic abstract is genuine, learned from 32 labelled training examples.

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.
Categoryreasoning
Subcategoryscientific-abstract keyword authenticity verification (binary), few-shot
Page statusunknown
Metricaccuracy
Directionhigher_is_better
Unit%
Dataset size2828
PublisherCLUE team

What it measures

An abstract from a Chinese academic paper plus a short list of keywords, some genuine and some fabricated by TF-IDF; the model judges whether every listed keyword is genuine, learned few-shot from 32 labelled training examples.

Task format

Binary keyword-authenticity classification, graded on the single correct label; evaluated from a 32-example few-shot training split, one of five parallel splits FewCLUE provides for this task.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub