Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSelf-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP
التشخيص الذاتي والدوائر الذاتية: اقتراح لتقليل التحيز القائم على Corpus في NLP
195
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
مجردة ⚠ تحتوي هذه الورقة على مطالبات ونواتج النماذج المسيئة في الطبيعة. عند التدريب على الزحف الكبيرة وغير المرفقة من الإنترنت، تلتقط نماذج اللغة وإعادة إنتاج جميع أنواع التحيزات غير المرغوب فيها التي يمكن العثور عليها في البيانات: أنها غالبا ما تولد لغة عنصرية أو جنسية أو عنيفة أو غير سامة. نظرا لأن النماذج الكبيرة تتطلب ملايين الأمثلة التدريبية لتحقيق أداء جيد، فمن الصعب منعها تماما من التعرض لمثل هذا المحتوى. في هذه الورقة، نوضح أولا في العثور على إيجاد مفاجئ: تعترف نماذج اللغة المحددة، إلى درجة كبيرة، تحيزاتهم غير المرغوب فيها وسمية المحتوى الذي ينتجونه. نشير إلى هذه القدرة كتشخيص الذاتي. بناء على هذا النتيجة، نقترح خوارزمية فك تشفير ذلك، بالنظر إلى وصف نصي فقط للسلوك غير المرغوب فيه، يقلل من احتمال إنتاج نموذج لغة ينتج نصا مشكلة. نشير إلى هذا النهج كدخل ذاتي. لا يعتمد الدخل الذاتي على قوائم Word يدويا يدويا، ولا يتطلب الأمر أي بيانات تدريبية أو تغييرات على معلمات النموذج. في حين أننا لا نقضاء بأي حال من الأحوال قضية نماذج اللغة التي تولد نص متحيز، فإننا نعتقد أن نهجنا خطوة مهمة في هذا الاتجاه
Abstract ⚠ This paper contains prompts and model outputs that are offensive in nature. When trained on large, unfiltered crawls from the Internet, language models pick up and reproduce all kinds of undesirable biases that can be found in the data: They often generate racist, sexist, violent, or otherwise toxic language. As large models require millions of training examples to achieve good performance, it is difficult to completely prevent them from being exposed to such content. In this paper, we first demonstrate a surprising finding: Pretrained language models recognize, to a considerable degree, their undesirable biases and the toxicity of the content they produce. We refer to this capability as self-diagnosis. Based on this finding, we then propose a decoding algorithm that, given only a textual description of the undesired behavior, reduces the probability of a language model producing problematic text. We refer to this approach as self-debiasing. Self-debiasing does not rely on manually curated word lists, nor does it require any training data or changes to the model's parameters. While we by no means eliminate the issue of language models generating biased text, we believe our approach to be an important step in this direction.1
References used
https://aclanthology.org/
rate research
Read More
Neural networks are the state-of-the-art method of machine learning for many problems in NLP. Their success in machine translation and other NLP tasks is phenomenal, but their interpretability is challenging. We want to find out how neural networks r
epresent meaning. In order to do this, we propose to examine the distribution of meaning in the vector space representation of words in neural networks trained for NLP tasks. Furthermore, we propose to consider various theories of meaning in the philosophy of language and to find a methodology that would enable us to connect these areas.
Exposure bias has been regarded as a central problem for auto-regressive language models (LM). It claims that teacher forcing would cause the test-time generation to be incrementally distorted due to the training-generation discrepancy. Although a lo
t of algorithms have been proposed to avoid teacher forcing and therefore alleviate exposure bias, there is little work showing how serious the exposure bias problem actually is. In this work, we focus on the task of open-ended language generation, propose metrics to quantify the impact of exposure bias in the aspects of quality, diversity, and consistency. Our key intuition is that if we feed ground-truth data prefixes (instead of prefixes generated by the model itself) into the model and ask it to continue the generation, the performance should become much better because the training-generation discrepancy in the prefix is removed. Both automatic and human evaluations are conducted in our experiments. On the contrary to the popular belief in exposure bias, we find that the the distortion induced by the prefix discrepancy is limited, and does not seem to be incremental during the generation. Moreover, our analysis reveals an interesting self-recovery ability of the LM, which we hypothesize to be countering the harmful effects from exposure bias.
Abstract Debugging a machine learning model is hard since the bug usually involves the training data and the learning process. This becomes even harder for an opaque deep learning model if we have no clue about how the model actually works. In this s
urvey, we review papers that exploit explanations to enable humans to give feedback and debug NLP models. We call this problem explanation-based human debugging (EBHD). In particular, we categorize and discuss existing work along three dimensions of EBHD (the bug context, the workflow, and the experimental setting), compile findings on how EBHD components affect the feedback providers, and highlight open problems that could be future research directions.
We analyze 6.7 million case law documents to determine the presence of gender bias within our judicial system. We find that current bias detection methods in NLP are insufficient to determine gender bias in our case law database and propose an altern
ative approach. We show that existing algorithms' inconsistent results are consequences of prior research's inconsistent definitions of biases themselves. Bias detection algorithms rely on groups of words to represent bias (e.g., salary,' job,' and boss' to represent employment as a potentially biased theme against women in text). However, the methods to build these groups of words have several weaknesses, primarily that the word lists are based on the researchers' own intuitions. We suggest two new methods of automating the creation of word lists to represent biases. We find that our methods outperform current NLP bias detection methods. Our research improves the capabilities of NLP technology to detect bias and highlights gender biases present in influential case law. In order to test our NLP bias detection method's performance, we regress our results of bias in case law against U.S census data of women's participation in the workforce in the last 100 years.
Open-domain chatbots are supposed to converse freely with humans without being restricted to a topic, task or domain. However, the boundaries and/or contents of open-domain conversations are not clear. To clarify the boundaries of openness'', we cond
uct two studies: First, we classify the types of speech events'' encountered in a chatbot evaluation data set (i.e., Meena by Google) and find that these conversations mainly cover the small talk'' category and exclude the other speech event categories encountered in real life human-human communication. Second, we conduct a small-scale pilot study to generate online conversations covering a wider range of speech event categories between two humans vs. a human and a state-of-the-art chatbot (i.e., Blender by Facebook). A human evaluation of these generated conversations indicates a preference for human-human conversations, since the human-chatbot conversations lack coherence in most speech event categories. Based on these results, we suggest (a) using the term small talk'' instead of open-domain'' for the current chatbots which are not that open'' in terms of conversational abilities yet, and (b) revising the evaluation methods to test the chatbot conversations against other speech events.
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
