clozelike

Clozelike is an adjective used to describe tasks, materials, or phenomena that resemble a cloze test, in which words are intentionally omitted from a passage and the reader supplies the missing words. The term derives from the cloze procedure, a technique developed in the 1950s and popularized in language testing to measure reading comprehension and lexical knowledge.

Variants of clozelike tasks differ in how omissions are selected and how many words are omitted. Common approaches include fixed-gap methods, where every nth word is removed, and random-gap methods, where omissions are scattered throughout the text. Scoring typically uses the proportion of correct completions or a partial-credit scale, depending on the precision required by the assessment.

Clozelike tasks are used in education and linguistics to assess reading comprehension, vocabulary, inference, and grammar. In language teaching, such exercises can target learners’ ability to infer meaning from context and to predict language form from semantic cues.

In natural language processing, the term clozelike is often applied to masked-language-modeling tasks that resemble cloze tests, where models predict missing tokens. This usage highlights the parallel between human completion of omissions and machine prediction of masked content. The design choices—masking strategy, omission patterns, and evaluation criteria—significantly influence difficulty and interpretability.

Example: The quick brown fox jumps over the lazy __.

See also: Cloze test, Cloze procedure, masked language modeling, fill-in-the-blank.