{"url":"/dataset/jampatoisnli","name":"JamPatoisNLI","full_name":null,"description_markdown":"JamPatoisNLI provides the first dataset for natural language inference in a creole language,\r\nJamaican Patois. Many of the most-spoken\r\nlow-resource languages are creoles. These\r\nlanguages commonly have a lexicon derived\r\nfrom a major world language and a distinctive grammar reflecting the languages of the\r\noriginal speakers and the process of language\r\nbirth by creolization. This gives them a distinctive place in exploring the effectiveness of\r\ntransfer from large monolingual or multilingual pretrained models. While our work, along\r\nwith previous work, shows that transfer from\r\nthese models to low-resource languages that\r\nare unrelated to languages in their training set\r\nis not very effective, we would expect stronger\r\nresults from transfer to creoles. Indeed, our\r\nexperiments show considerably better results\r\nfrom few-shot learning of JamPatoisNLI than\r\nfor such unrelated languages, and help us begin to understand how the unique relationship\r\nbetween creoles and their high-resource base\r\nlanguages affect cross-lingual transfer. JamPatoisNLI, which consists of naturally-occurring\r\npremises and expert-written hypotheses, is a\r\nstep towards steering research into a traditionally underserved language and a useful benchmark for understanding cross-lingual NLP.","description_withheld":null,"homepage":"","introduced_date":null,"introduced_date_note":null,"introduced_by":null,"license":null,"modalities":[],"tasks":[{"name":"Natural Language Inference","url":"/task/natural-language-inference","datasets_with_task":"/datasets/task/natural-language-inference"}],"languages":[{"name":"Jamaican Creole English","url":"/datasets/language/jamaican-creole-english"}],"variants":["JamPatoisNLI"],"data_loaders":[],"num_papers_in_archive":1,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/natural-language-inference-on-jampatoisnli","task":"Natural Language Inference","dataset_variant":"JamPatoisNLI","rows":2,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"roberta-unfrozen","paper":"/paper/jampatoisnli-a-jamaican-patois-natural","metrics":{"Accuracy":"76.50"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/jampatoisnli-a-jamaican-patois-natural","title":"JamPatoisNLI: A Jamaican Patois Natural Language Inference Dataset","date":"2022-12-07","rows_on_this_dataset":2,"code_links":0,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}