dynabench/qa

Tasks

• question-answering

Citation text from the card

@article{bartolo2020beat,
    author = {Bartolo, Max and Roberts, Alastair and Welbl, Johannes and Riedel, Sebastian and Stenetorp, Pontus},
    title = {Beat the AI: Investigating Adversarial Human Annotation for Reading Comprehension},
    journal = {Transactions of the Association for Computational Linguistics},
    volume = {8},
    number = {},
    pages = {662-678},
    year = {2020},
    doi = {10.1162/tacl\_a\_00338},
    URL = { https://doi.org/10.1162/tacl_a_00338 },
    eprint = { https://doi.org/10.1162/tacl_a_00338 },
    abstract = { Innovations in annotation methodology have been a catalyst for Reading Comprehension (RC) datasets and models. One recent trend to challenge current RC models is to involve a model in the annotation process: Humans create questions adversarially, such that the model fails to answer them correctly. In this work we investigate this annotation methodology and apply it in three different settings, collecting a total of 36,000 samples with progressively stronger models in the annotation loop. This allows us to explore questions such as the reproducibility of the adversarial effect, transfer from data collected with varying model-in-the-loop strengths, and generalization to data collected without a model. We find that training on adversarially collected samples leads to strong generalization to non-adversarially collected datasets, yet with progressive performance deterioration with increasingly stronger models-in-the-loop. Furthermore, w

All datasets marked Permissive or share-alike | JSON