GroundCocoa

Five-way multiple-choice flight-booking task that tests compositional and conditional reasoning over user constraints; 4,849 public test items.

Also known as: GroundCocoa, ground_cocoa, harsh147/GroundCocoa

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
Subcategorycompositional and conditional reasoning via five-way flight-option matching
Page statusactive
Metricaccuracy
Directionhigher_is_better
Unit%
Dataset size4849
Dataset licenceCC-BY-4.0
PublisherOhio State University NLP Group

What it measures

GroundCocoa asks a model to pick which of five scraped flight options satisfies a long English user request. Requests are built from 2–6 flight slots (airline, times, price, layovers, carbon, and others) combined as a product-of-sums formula, then paraphrased with GPT-4 Turbo and checked by hand. The skill is if-then and multi-constraint grounding, not dialogue state tracking and not a travel agent with tools. Atypical queries include odd wants such as more layovers or above-average carbon.

Task format

Five-way multiple choice. lm-evaluation-harness task groundcocoa formats the query and options A–E and scores log-likelihood of "The answer is Option {A–E}". output_type multiple_choice. metric acc. Paper also reports zero-shot, chain-of-thought (full and partial), and least-to-most prompting. L2M numbers in the paper use a 200-item subset.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub