Dark Humor Detection (BIG-bench)

80-item BIG-bench binary task: decide whether a short English text is intended as a dark joke.

Also known as: Dark Humor Detection

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
SubcategoryBIG-bench binary dark-joke vs not-a-joke classification
Page statusunknown
Metricmultiple_choice_grade
Directionhigher_is_better
Unit%
Dataset size80
Dataset licenceCC0-1.0 (task README); Apache-2.0 (google/BIG-bench repository)
PublisherSaarland University / DFKI (via the BIG-bench collaboration)

What it measures

dark_humor_detection shows one English snippet and asks whether it is intended as a dark joke. Gold labels are joke versus not a joke, 40 each in task.json. Non-jokes are punchlines rewritten to similar length. The authors (Saarland/DFKI) say the target is intent, not funniness. Dummy-model header: 80 multiple-choice queries. Not in [BIG-Bench Hard](bbh.md).

Task format

JSON multiple-choice, preferred_score multiple_choice_grade. task_prefix asks for "joke" or "not a joke". append_choices_to_input is false. Canary GUID embedded. lm-eval ships bigbench_dark_humor_detection_multiple_choice and bigbench_dark_humor_detection_generate_until on hails/bigbench, dataset_name dark_humor_detection_zero_shot.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub