DISFL-QA

An 8,000-item BIG-bench task built by inserting realistic speech disfluencies into SQuAD v2 questions, to test whether models stay robust when a question corrects or restarts itself.

Also known as: Disfl-QA, DISFL-QA: A Benchmark Dataset for Understanding Disfluencies in Question Answering

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
Subcategoryextractive reading comprehension under disfluent (self-corrected, restarted) questions
Page statusactive
Metricexact_str_match
Directionhigher_is_better
Unit%
Dataset size8000
Dataset licenceCC BY 4.0
PublisherGoogle Research (submitted to the BIG-bench collaboration)

What it measures

disfl_qa asks a model to pick the correct answer span from a Wikipedia passage, given a question that has been rewritten to contain a natural contextual disfluency -- a correction or restart such as "What part [of] the composition of the Earth's biosphere is comprised of oxygen no sorry the Earth's atmosphere?" It measures robustness to a specific, understudied failure mode: whether shallow question-context matching breaks down once the question itself contains noise a fluent reader would filter out. Over 90% of DISFL-QA's disfluencies are corrections or restarts rather than simple filler, which the original paper's authors built specifically to be harder than earlier disfluency datasets that mostly repeated words.

Task format

Extractive question answering: given a disfluent question and a context passage, output the answer span (or "unknown" for unanswerable items); scored by exact_str_match against a single gold reference. A BIG-bench canary GUID is embedded in the task file.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub