Pith. sign in

REVIEW 1 cited by

Text Detoxification using Large Pre-trained Neural Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2109.08914 v2 pith:IPDSZVJS submitted 2021-09-18 cs.CL cs.LG

classification cs.CLcs.LG
keywords modelsstyletransfermethodmethodstexttoxicitybert
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present two novel unsupervised methods for eliminating toxicity in text. Our first method combines two recent ideas: (1) guidance of the generation process with small style-conditional language models and (2) use of paraphrasing models to perform style transfer. We use a well-performing paraphraser guided by style-trained language models to keep the text content and remove toxicity. Our second method uses BERT to replace toxic words with their non-offensive synonyms. We make the method more flexible by enabling BERT to replace mask tokens with a variable number of words. Finally, we present the first large-scale comparative study of style transfer models on the task of toxicity removal. We compare our models with a number of methods for style transfer. The models are evaluated in a reference-free way using a combination of unsupervised style transfer metrics. Both methods we suggest yield new SOTA results.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Generative Data Refinement: Just Ask for Better Data

    cs.LG 2025-09 conditional novelty 4.0 of 10

    A pretrained LLM can rewrite individual data samples to strip out PII or toxic content while preserving useful information, creating safer training data.

Pith tools