Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userMeasuring Social Biases in Grounded Vision and Language Embeddings
قياس التحيزات الاجتماعية في الرؤية الأساسية وإدماج اللغة
372
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
نحن نعمل على تعميم فكرة قياس التحيزات الاجتماعية في Word Ageddings لإضاءة Word بصريا. الحياز موجودة في المدينات المتطرفة، ويبدو أنها بالفعل أكثر أهمية أو أكثر أهمية من المدمج غير المقصود. هذا على الرغم من حقيقة أن الرؤية واللغة يمكن أن تعاني من تحيزات مختلفة، والذي قد يأمل المرء أن يخفف من التحيزات في كليهما. توجد طرق متعددة لتعميم التحيز القياسي القياسي في Word Ageddings لهذا الإعداد الجديد. نقدم مساحة التعميمات (Weat-Weat-Weat and Grounded) وإظهار أن ثلاث تعميمات تجيب على أسئلة مختلفة لكنها مهمة حول كيفية تفاعل التحيزات واللغة والرؤية. يتم استخدام هذه المقاييس في مجموعة بيانات جديدة، الأول من أجل التحيز الأساسي، الذي تم إنشاؤه عن طريق زيادة معايير التحيز اللغوي القياسي مع 10228 صورة من كوكو، والتسمية التوضيحية المفاهيمية، وصور جوجل. بناء البيانات يتحدى لأن مجموعات بيانات الرؤية هي نفسها منحازة للغاية. سيبدأ وجود هذه التحيزات في الأنظمة في الحصول على عواقب عالمية حقيقية حيث يتم نشرها، مما يجعلها تقيس التحيز بعناية ثم تخفيفها بالغ الأهمية لبناء مجتمع عادل.
We generalize the notion of measuring social biases in word embeddings to visually grounded word embeddings. Biases are present in grounded embeddings, and indeed seem to be equally or more significant than for ungrounded embeddings. This is despite the fact that vision and language can suffer from different biases, which one might hope could attenuate the biases in both. Multiple ways exist to generalize metrics measuring bias in word embeddings to this new setting. We introduce the space of generalizations (Grounded-WEAT and Grounded-SEAT) and demonstrate that three generalizations answer different yet important questions about how biases, language, and vision interact. These metrics are used on a new dataset, the first for grounded bias, created by augmenting standard linguistic bias benchmarks with 10,228 images from COCO, Conceptual Captions, and Google Images. Dataset construction is challenging because vision datasets are themselves very biased. The presence of these biases in systems will begin to have real-world consequences as they are deployed, making carefully measuring bias and then mitigating it critical to building a fair society.
References used
https://aclanthology.org/
rate research
Read More
Biases continue to be prevalent in modern text and media, especially subjective bias -- a special type of bias that introduces improper attitudes or presents a statement with the presupposition of truth. To tackle the problem of detecting and further
mitigating subjective bias, we introduce a manually annotated parallel corpus WIKIBIAS with more than 4,000 sentence pairs from Wikipedia edits. This corpus contains annotations towards both sentence-level bias types and token-level biased segments. We present systematic analyses of our dataset and results achieved by a set of state-of-the-art baselines in terms of three tasks: bias classification, tagging biased segments, and neutralizing biased text. We find that current models still struggle with detecting multi-span biases despite their reasonable performances, suggesting that our dataset can serve as a useful research benchmark. We also demonstrate that models trained on our dataset can generalize well to multiple domains such as news and political speeches.
Abstract Measuring bias is key for better understanding and addressing unfairness in NLP/ML models. This is often done via fairness metrics, which quantify the differences in a model's behaviour across a range of demographic groups. In this work, we
shed more light on the differences and similarities between the fairness metrics used in NLP. First, we unify a broad range of existing metrics under three generalized fairness metrics, revealing the connections between them. Next, we carry out an extensive empirical comparison of existing metrics and demonstrate that the observed differences in bias measurement can be systematically explained via differences in parameter choices for our generalized metrics.
Phrase grounding aims to map textual phrases to their associated image regions, which can be a prerequisite for multimodal reasoning and can benefit tasks requiring identifying objects based on language. With pre-trained vision-and-language models ac
hieving impressive performance across tasks, it remains unclear if we can directly utilize their learned embeddings for phrase grounding without fine-tuning. To this end, we propose a method to extract matched phrase-region pairs from pre-trained vision-and-language embeddings and propose four fine-tuning objectives to improve the model phrase grounding ability using image-caption data without any supervised grounding signals. Experiments on two representative datasets demonstrate the effectiveness of our objectives, outperforming baseline models in both weakly-supervised and supervised phrase grounding settings. In addition, we evaluate the aligned embeddings on several other downstream tasks and show that we can achieve better phrase grounding without sacrificing representation generality.
Word embeddings are widely used in Natural Language Processing (NLP) for a vast range of applications. However, it has been consistently proven that these embeddings reflect the same human biases that exist in the data used to train them. Most of the
introduced bias indicators to reveal word embeddings' bias are average-based indicators based on the cosine similarity measure. In this study, we examine the impacts of different similarity measures as well as other descriptive techniques than averaging in measuring the biases of contextual and non-contextual word embeddings. We show that the extent of revealed biases in word embeddings depends on the descriptive statistics and similarity measures used to measure the bias. We found that over the ten categories of word embedding association tests, Mahalanobis distance reveals the smallest bias, and Euclidean distance reveals the largest bias in word embeddings. In addition, the contextual models reveal less severe biases than the non-contextual word embedding models.
Automatic detection of toxic language plays an essential role in protecting social media users, especially minority groups, from verbal abuse. However, biases toward some attributes, including gender, race, and dialect, exist in most training dataset
s for toxicity detection. The biases make the learned models unfair and can even exacerbate the marginalization of people. Considering that current debiasing methods for general natural language understanding tasks cannot effectively mitigate the biases in the toxicity detectors, we propose to use invariant rationalization (InvRat), a game-theoretic framework consisting of a rationale generator and a predictor, to rule out the spurious correlation of certain syntactic patterns (e.g., identity mentions, dialect) to toxicity labels. We empirically show that our method yields lower false positive rate in both lexical and dialectal attributes than previous debiasing methods.
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
