ParaDetox: Detoxification with Parallel Data (English). Toxicity Task Results

This repository contains information about Toxicity Task markup from English Paradetox dataset collection pipeline. The original paper "ParaDetox: Detoxification with Parallel Data" was presented at ACL 2022 main conference.

ParaDetox Collection Pipeline

The ParaDetox Dataset collection was done via Yandex.Toloka crowdsource platform. The collection was done in three steps:

• Task 1: Generation of Paraphrases : The first crowdsourcing task asks users to eliminate toxicity in a given sentence while keeping the content.
• Task 2: Content Preservation Check : We show users the generated paraphrases along with their original variants and ask them to indicate if they have close meanings.
• Task 3: Toxicity Check : Finally, we check if the workers succeeded in removing toxicity.

Specifically this repo contains the results of Task 3: Toxicity Check . Here, the samples with markup confidence >= 90 are present. The input here is text and the label shows if the text is toxic or not. Totally, datasets contains 26,507 samples. Among them, the minor part is toxic examples (4,009 pairs).

Citation

@inproceedings{logacheva-etal-2022-paradetox,
    title = "{P}ara{D}etox: Detoxification with Parallel Data",
    author = "Logacheva, Varvara  and
      Dementieva, Daryna  and
      Ustyantsev, Sergey  and
      Moskovskiy, Daniil  and
      Dale, David  and
      Krotova, Irina  and
      Semenov, Nikita  and
      Panchenko, Alexander",
    booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = may,
    year = "2022",
    address = "Dublin, Ireland",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2022.acl-long.469",
    pages = "6804--6818",
    abstract = "We present a novel pipeline for the collection of parallel data for the detoxification task. We collect non-toxic paraphrases for over 10,000 English toxic sentences. We also show that this pipeline can be used to distill a large existing corpus of paraphrases to get toxic-neutral sentence pairs. We release two parallel corpora which can be used for the training of detoxification models. To the best of our knowledge, these are the first parallel datasets for this task.We describe our pipeline in detail to make it fast to set up for a new language or domain, thus contributing to faster and easier development of new parallel resources.We train several detoxification models on the collected data and compare them with several baselines and state-of-the-art unsupervised approaches. We conduct both automatic and manual evaluations. All models trained on parallel data outperform the state-of-the-art unsupervised models by a large margin. This suggests that our novel datasets can boost the performance of detoxification systems.",
}

Contacts

For any questions, please contact: Daryna Dementieva ( dardem96@gmail.com )