s-nlp/pseudoparadetox_llama3_8b_10shot_patched

Dataset Card for Dataset Name

Dataset Details

Using this data, we got the highlighted results using BART sequence-to-sequence model. The configs and code for fine-tuning can be found on github

Dataset Description

This is a PseudoParaDetox dataset with real source toxic data and generated neutral detoxification by a patched LLama 3 8B with 10-shot. This dataset is based on the ParaDetox dataset for English texts detoxification.

• Curated by: Daniil Moskovskiy, Sergey Pletenev
• Language(s) (NLP): [English]
• License: [OpenRAIL]

Dataset Sources [optional]

• Repository: Link
• Paper: Link
• Website: Link

Uses

Direct Use

This dataset can be used for fine-tuning text detoxification models.

Out-of-Scope Use

Dataset Structure

[More Information Needed]

Dataset Creation

Curation Rationale

[More Information Needed]

Source Data

Data Collection and Processing

[More Information Needed]

Bias, Risks, and Limitations

Some parts of this data can be found offensive for several

Recommendations

Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.

Citation

If you use our data, please cite our paper.

BibTeX:

@inproceedings{moskovskiy-etal-2024-llms,
    title = "{LLM}s to Replace Crowdsourcing For Parallel Data Creation? The Case of Text Detoxification",
    author = "Moskovskiy, Daniil  and
      Pletenev, Sergey  and
      Panchenko, Alexander",
    editor = "Al-Onaizan, Yaser  and
      Bansal, Mohit  and
      Chen, Yun-Nung",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
    month = nov,
    year = "2024",
    address = "Miami, Florida, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.findings-emnlp.839/",
    doi = "10.18653/v1/2024.findings-emnlp.839",
    pages = "14361--14373",
    abstract = "The lack of high-quality training data remains a significant challenge in NLP. Manual annotation methods, such as crowdsourcing, are costly, require intricate task design skills, and, if used incorrectly, may result in poor data quality. From the other hand, LLMs have demonstrated proficiency in many NLP tasks, including zero-shot and few-shot data annotation. However, they often struggle with text detoxification due to alignment constraints and fail to generate the required detoxified text. This work explores the potential of modern open source LLMs to annotate parallel data for text detoxification. Using the recent technique of activation patching, we generate a pseudo-parallel detoxification dataset based on ParaDetox. The detoxification model trained on our generated data shows comparable performance to the original dataset in automatic detoxification evaluation metrics and superior quality in manual evaluation and side-by-side comparisons."
}

APA:

Moskovskiy, D., Pletenev, S., & Panchenko, A. (2024, November). LLMs to Replace Crowdsourcing For Parallel Data Creation? The Case of Text Detoxification. In Findings of the Association for Computational Linguistics: EMNLP 2024 (pp. 14361-14373).

Dataset Card Contact

Daniil Moskovskiy