Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDetoxifying Language Models Risks Marginalizing Minority Voices
نماذج لغة إزالة السموم مخاطر تهميش أصوات الأقلية
267
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
يجب أن تكون نماذج اللغة (LMS) آمنة وعادلة أن تكون مرتدة في الممارسة العملية.مع مراعاة السلامة، تم اقتراح العديد من تقنيات إزالة السموم (E.G.، Dathathri et al. 2020؛ Krause et al. 2020) تخفيف أجيال LM السامة.في هذا العمل، نظير على أن تقنيات إزالة السموم هذه تؤذي الأسهم: أنها تقلل من فائدة LMS على اللغة التي تستخدمها المجموعات المهمشة (على سبيل المثال، إن العربية الأفريقية الأمريكية وهوية الهوية الأقلية).على وجه الخصوص، نقوم بتنفيذ التقييمات التلقائية والإنسانية لجودة جيل النص عندما تكون LMS مشروطة على المدخلات مع لهجات مختلفة ومعرفات المجموعة.نجد أن إزالة السموم يجعل LMS أكثر هشة من تحول التوزيع، خاصة في اللغة التي تستخدمها المجموعات المهمشة.نحدد أن هذه الإخفاقات تنبع من أساليب إزالة السموم التي تستغل ارتباطات زائفة في مجموعات بيانات السمية.بشكل عام، تسلط نتائجنا الضوء على التوتر بين إمكانية التحكم والموازنة التوزيعية ل LMS.
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 these 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.
References used
https://aclanthology.org/
rate research
Read More
Large language models (LM) generate remarkably fluent text and can be efficiently adapted across NLP tasks. Measuring and guaranteeing the quality of generated text in terms of safety is imperative for deploying LMs in the real world; to this end, pr
ior work often relies on automatic evaluation of LM toxicity. We critically discuss this approach, evaluate several toxicity mitigation strategies with respect to both automatic and human evaluation, and analyze consequences of toxicity mitigation in terms of model bias and LM quality. We demonstrate that while basic intervention strategies can effectively optimize previously established automatic metrics on the REALTOXICITYPROMPTS dataset, this comes at the cost of reduced LM coverage for both texts about, and dialects of, marginalized groups. Additionally, we find that human raters often disagree with high automatic toxicity scores after strong toxicity reduction interventions---highlighting further the nuances involved in careful evaluation of LM toxicity.
Abstract We present a language model that combines a large parametric neural network (i.e., a transformer) with a non-parametric episodic memory component in an integrated architecture. Our model uses extended short-term context by caching local hidd
en states---similar to transformer-XL---and global long-term memory by retrieving a set of nearest neighbor tokens at each timestep. We design a gating function to adaptively combine multiple information sources to make a prediction. This mechanism allows the model to use either local context, short-term memory, or long-term memory (or any combination of them) on an ad hoc basis depending on the context. Experiments on word-based and character-based language modeling datasets demonstrate the efficacy of our proposed method compared to strong baselines.
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.
Pre-trained multilingual language models have become an important building block in multilingual Natural Language Processing. In the present paper, we investigate a range of such models to find out how well they transfer discourse-level knowledge acr
oss languages. This is done with a systematic evaluation on a broader set of discourse-level tasks than has been previously been assembled. We find that the XLM-RoBERTa family of models consistently show the best performance, by simultaneously being good monolingual models and degrading relatively little in a zero-shot setting. Our results also indicate that model distillation may hurt the ability of cross-lingual transfer of sentence representations, while language dissimilarity at most has a modest effect. We hope that our test suite, covering 5 tasks with a total of 22 languages in 10 distinct families, will serve as a useful evaluation platform for multilingual performance at and beyond the sentence level.
Pretrained multilingual language models have become a common tool in transferring NLP capabilities to low-resource languages, often with adaptations. In this work, we study the performance, extensibility, and interaction of two such adaptations: voca
bulary augmentation and script transliteration. Our evaluations on part-of-speech tagging, universal dependency parsing, and named entity recognition in nine diverse low-resource languages uphold the viability of these approaches while raising new questions around how to optimally adapt multilingual models to low-resource settings.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
