Added Toxicity Mitigation at Inference Time for Multimodal and Massively Multilingual Translation

Marta R. Costa-jussà, David Dale, Maha Elbayad, Bokai Yu
11/11/2023
cs.CL

No confident code implementation found for this paper.

Abstract

Added toxicity in the context of translation refers to the fact of producing a translation output with more toxicity than there exists in the input. In this paper, we present MinTox which is a novel pipeline to identify added toxicity and mitigate this issue which works at inference time. MinTox uses a toxicity detection classifier which is multimodal (speech and text) and works in languages at scale. The mitigation method is applied to languages at scale and directly in text outputs. MinTox is applied to SEAMLESSM4T, which is the latest multimodal and massively multilingual machine translation system. For this system, MinTox achieves significant added toxicity mitigation across domains, modalities and language directions. MinTox manages to approximately filter out from 25% to 95% of added toxicity (depending on the modality and domain) while keeping translation quality.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Translate, then Detect: Leveraging Machine Translation for Cross-Lingual Toxicity Classification

90% match2025

ReSeTOX: Re-learning attention weights for toxicity mitigation in machine translation

89% match2023

A Survey of Toxicity Detection and Mitigation Strategies for Multilingual Language Models

86% match2026

From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models

85% match1 repo252024

PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models

84% match2 repos1322024

Cross-Domain Toxic Spans Detection

83% match2023

Cite this paper

@article{costajuss2023added,
  title  = {Added Toxicity Mitigation at Inference Time for Multimodal and Massively Multilingual Translation},
  author = {Marta R. Costa-jussà and David Dale and Maha Elbayad and Bokai Yu},
  year   = {2023},
  eprint = {2311.06532},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2311.06532v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2311.06532.svg)](https://codeofpaper.com/paper/2311.06532)

Discussion