Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userLow Frequency Names Exhibit Bias and Overfitting in Contextualizing Language Models
تظهر أسماء التردد المنخفض التحيز والتجول في نماذج اللغة السياقية
311
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
نستخدم مجموعة بيانات من الأسماء الأولى الأمريكية مع ملصقات تستند إلى النوع الاجتماعي السائد والمجموعة العرقية لفحص تأثير تواتر Corpus على التقييم والسياق والتشابه إلى التمثيل الأولي والتحيز في Bert و GPT-2 و T5 و XLNet. نظهر أن الأسماء الأكثر في الغالب والأسماء غير البيضاء أقل تواترا في شركة التدريب لهذه النماذج الأربع هذه. نجد أن الأسماء النادرة هي أكثر مماثلة ذاتيا عبر السياقات، مع Rho Spearman بين التردد والتشابه الذاتي بنسبة منخفضة تصل إلى 763. الأسماء النادرة هي أيضا أقل تشبه التمثيل الأولي، مع تشابه RHO ل Spearman بين التردد ومحاذاة النواة الخطية (CKA) للتمثيل الأولي بما يصل إلى .702. علاوة على ذلك، نجد Rho Spearman بين التحيز العنصري وتكرار الاسم في Bert of .492، مما يشير إلى أن أسماء مجموعات الأقليات ذات التردد الأدنى مرتبطون ببراعة. تخضع تمثيل الأسماء النادرة لمعالجة المزيد من المعالجة، ولكنها أكثر مماثلة ذاتيا، مما يشير إلى أن النماذج تعتمد على تمثيل أقل مستنيرة في السياق بأسماء غير شائعة وأسماء الأقليات التي يتم إجاءاتها على عدد أقل من السياقات الملحوظة.
We use a dataset of U.S. first names with labels based on predominant gender and racial group to examine the effect of training corpus frequency on tokenization, contextualization, similarity to initial representation, and bias in BERT, GPT-2, T5, and XLNet. We show that predominantly female and non-white names are less frequent in the training corpora of these four language models. We find that infrequent names are more self-similar across contexts, with Spearman's rho between frequency and self-similarity as low as -.763. Infrequent names are also less similar to initial representation, with Spearman's rho between frequency and linear centered kernel alignment (CKA) similarity to initial representation as high as .702. Moreover, we find Spearman's rho between racial bias and name frequency in BERT of .492, indicating that lower-frequency minority group names are more associated with unpleasantness. Representations of infrequent names undergo more processing, but are more self-similar, indicating that models rely on less context-informed representations of uncommon and minority names which are overfit to a lower number of observed contexts.
References used
https://aclanthology.org/
rate research
Read More
Taxonomies are symbolic representations of hierarchical relationships between terms or entities. While taxonomies are useful in broad applications, manually updating or maintaining them is labor-intensive and difficult to scale in practice. Conventio
nal supervised methods for this enrichment task fail to find optimal parents of new terms in low-resource settings where only small taxonomies are available because of overfitting to hierarchical relationships in the taxonomies. To tackle the problem of low-resource taxonomy enrichment, we propose Musubu, an efficient framework for taxonomy enrichment in low-resource settings with pretrained language models (LMs) as knowledge bases to compensate for the shortage of information. Musubu leverages an LM-based classifier to determine whether or not inputted term pairs have hierarchical relationships. Musubu also utilizes Hearst patterns to generate queries to leverage implicit knowledge from the LM efficiently for more accurate prediction. We empirically demonstrate the effectiveness of our method in extensive experiments on taxonomies from both a SemEval task and real-world retailer datasets.
One of the central aspects of contextualised language models is that they should be able to distinguish the meaning of lexically ambiguous words by their contexts. In this paper we investigate the extent to which the contextualised embeddings of word
forms that display multiplicity of sense reflect traditional distinctions of polysemy and homonymy. To this end, we introduce an extended, human-annotated dataset of graded word sense similarity and co-predication acceptability, and evaluate how well the similarity of embeddings predicts similarity in meaning. Both types of human judgements indicate that the similarity of polysemic interpretations falls in a continuum between identity of meaning and homonymy. However, we also observe significant differences within the similarity ratings of polysemes, forming consistent patterns for different types of polysemic sense alternation. Our dataset thus appears to capture a substantial part of the complexity of lexical ambiguity, and can provide a realistic test bed for contextualised embeddings. Among the tested models, BERT Large shows the strongest correlation with the collected word sense similarity ratings, but struggles to consistently replicate the observed similarity patterns. When clustering ambiguous word forms based on their embeddings, the model displays high confidence in discerning homonyms and some types of polysemic alternations, but consistently fails for others.
Relational knowledge bases (KBs) are commonly used to represent world knowledge in machines. However, while advantageous for their high degree of precision and interpretability, KBs are usually organized according to manually-defined schemas, which l
imit their expressiveness and require significant human efforts to engineer and maintain. In this review, we take a natural language processing perspective to these limitations, examining how they may be addressed in part by training deep contextual language models (LMs) to internalize and express relational knowledge in more flexible forms. We propose to organize knowledge representation strategies in LMs by the level of KB supervision provided, from no KB supervision at all to entity- and relation-level supervision. Our contributions are threefold: (1) We provide a high-level, extensible taxonomy for knowledge representation in LMs; (2) Within our taxonomy, we highlight notable models, evaluation tasks, and findings, in order to provide an up-to-date review of current knowledge representation capabilities in LMs; and (3) We suggest future research directions that build upon the complementary aspects of LMs and KBs as knowledge representations.
The success of large-scale contextual language models has attracted great interest in probing what is encoded in their representations. In this work, we consider a new question: to what extent contextual representations of concrete nouns are aligned
with corresponding visual representations? We design a probing model that evaluates how effective are text-only representations in distinguishing between matching and non-matching visual representations. Our findings show that language representations alone provide a strong signal for retrieving image patches from the correct object categories. Moreover, they are effective in retrieving specific instances of image patches; textual context plays an important role in this process. Visually grounded language models slightly outperform text-only language models in instance retrieval, but greatly under-perform humans. We hope our analyses inspire future research in understanding and improving the visual capabilities of language models.
Nowadays, social media platforms use classification models to cope with hate speech and abusive language. The problem of these models is their vulnerability to bias. A prevalent form of bias in hate speech and abusive language datasets is annotator b
ias caused by the annotator's subjective perception and the complexity of the annotation task. In our paper, we develop a set of methods to measure annotator bias in abusive language datasets and to identify different perspectives on abusive language. We apply these methods to four different abusive language datasets. Our proposed approach supports annotation processes of such datasets and future research addressing different perspectives on the perception of abusive language.
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
