Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userChallenges in Detoxifying Language Models
التحديات في نماذج لغة السموم
509
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
تولد نماذج اللغة الكبيرة (LM) نص بطلاقة بشكل ملحوظ ويمكن تكييفها بكفاءة عبر مهام NLP. قياس وضمان جودة النص الذي تم إنشاؤه من حيث السلامة أمر ضروري لنشر LMS في العالم الحقيقي؛ تحقيقا لهذه الغاية، غالبا ما يعتمد العمل السابق على التقييم التلقائي لسمية LM. نناقش هذا النهج بشكل خطير، وتقييم العديد من استراتيجيات تخفيف السمية فيما يتعلق بالتقييم التلقائي والبشري، وتحليل عواقب التخفيف من السمية من حيث التحيز النموذجي وجودة LM. نوضح أنه في حين أن استراتيجيات التدخل الأساسية يمكن أن تتحسن بشكل فعال مقاييس تلقائية تم تأسيسها مسبقا على مجموعة بيانات Realtoxicyprompts، فإن هذا يأتي عند تكلفة انخفاض تغطية LM لكلا النصوص حول، ولهجات المجموعات المهمشة. بالإضافة إلى ذلك، نجد أن التصدير البشري غالبا ما يختلفون في درجات سمية تلقائية عالية بعد تدخلات تخفيض السمية القوي --- تسليط الضوء على مزيد من الفروق الدقيقة المشاركة في التقييم الدقيق لسامة LM.
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, prior 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.
References used
https://aclanthology.org/
rate research
Read More
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 generation
s. 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.
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 probe pre-trained transformer language models for bridging inference. We first investigate individual attention heads in BERT and observe that attention heads at higher layers prominently focus on bridging relations in-comparison with the lower an
d middle layers, also, few specific attention heads concentrate consistently on bridging. More importantly, we consider language models as a whole in our second approach where bridging anaphora resolution is formulated as a masked token prediction task (Of-Cloze test). Our formulation produces optimistic results without any fine-tuning, which indicates that pre-trained language models substantially capture bridging inference. Our further investigation shows that the distance between anaphor-antecedent and the context provided to language models play an important role in the inference.
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.
Recent progress in language modeling has been driven not only by advances in neural architectures, but also through hardware and optimization improvements. In this paper, we revisit the neural probabilistic language model (NPLM) of Bengio et al. (200
3), which simply concatenates word embeddings within a fixed window and passes the result through a feed-forward network to predict the next word. When scaled up to modern hardware, this model (despite its many limitations) performs much better than expected on word-level language model benchmarks. Our analysis reveals that the NPLM achieves lower perplexity than a baseline Transformer with short input contexts but struggles to handle long-term dependencies. Inspired by this result, we modify the Transformer by replacing its first self-attention layer with the NPLM's local concatenation layer, which results in small but consistent perplexity decreases across three word-level language modeling datasets.
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
