pith. sign in

arxiv: 2104.06390 · v1 · pith:7DKKB7LWnew · submitted 2021-04-13 · 💻 cs.CL · cs.LG

Detoxifying Language Models Risks Marginalizing Minority Voices

classification 💻 cs.CL cs.LG
keywords detoxificationlanguagegroupsmarginalizedminoritymodelstechniquesused
0
0 comments X
read the original abstract

Language models (LMs) must be both safe and equitable to be responsibly deployed in practice. With safety in mind, numerous detoxification techniques (e.g., Dathathri et al. 2020; Krause et al. 2020) have been proposed to mitigate toxic LM generations. In this work, we show that current detoxification techniques hurt equity: they decrease the utility of LMs on language used by marginalized groups (e.g., African-American English and minority identity mentions). In particular, we perform automatic and human evaluations of text generation quality when LMs are conditioned on inputs with different dialects and group identifiers. We find that detoxification makes LMs more brittle to distribution shift, especially on language used by marginalized groups. We identify that these failures stem from detoxification methods exploiting spurious correlations in toxicity datasets. Overall, our results highlight the tension between the controllability and distributional robustness of LMs.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

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

  1. Red Teaming Language Models with Language Models

    cs.CL 2022-02 conditional novelty 7.0

    One language model can generate diverse test cases to automatically uncover tens of thousands of harmful behaviors, including offensive replies and privacy leaks, in a large target language model.

  2. On The Effectiveness-Fluency Trade-Off In LLM Conditioning: A Systematic Study

    cs.CL 2026-06 unverdicted novelty 6.0

    Systematic experiments reveal that activation steering trades fluency for concept control, is less effective on instruction-tuned models, and that prompting/SFT excel at injection but not removal, with textual metrics...

  3. Ethical and social risks of harm from Language Models

    cs.CL 2021-12 accept novelty 6.0

    The authors provide a detailed taxonomy of 21 risks associated with language models, covering discrimination, information leaks, misinformation, malicious applications, interaction harms, and societal impacts like job...

  4. PaLM 2 Technical Report

    cs.CL 2023-05 unverdicted novelty 5.0

    PaLM 2 reports state-of-the-art results on language, reasoning, and multilingual tasks with improved efficiency over PaLM.