Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userHow Does Counterfactually Augmented Data Impact Models for Social Computing Constructs?
كيف تعزز نماذج تأثير البيانات المعزز بشكل مضاد لبناء الحوسبة الاجتماعية؟
245
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
نظرا لأن نماذج NLP منتشرة بشكل متزايد في إعدادات موقع اجتماعي مثل الكشف عن المحتوى المسيء عبر الإنترنت، فمن الضروري التأكد من أن هذه النماذج قوية. تتمثل إحدى طرق تحسين متواضع النموذج في إنشاء بيانات معدية مضادة للعمل (CAD) لنماذج التدريب التي يمكن أن تتعلم بشكل أفضل التمييز بين الميزات الأساسية وتذكير البيانات. في حين أن النماذج المدربة على هذا النوع من البيانات قد أظهرت تعميم تعميمات خارج نطاق، فلا تزال غير واضحة ما هي مصادر هذه التحسينات. نحقق في فوائد CAD لنماذج NLP الاجتماعية من خلال التركيز على ثلاثة بنيات الحوسبة الاجتماعية --- الشعور والجوعي، خطاب الكراهية. تقييم أداء النماذج المدربة مع وبدون CAD عبر أنواع مختلفة من مجموعات البيانات، نجد أنه أثناء قيام النماذج المدربة على CAD بإظهار الأداء السفلي داخل المجال، فإنهم يعممون أفضل من المجال. نحن نفث هذا التناقض الواضح باستخدام تفسيرات الجهاز والعثور على أن CAD يقلل من الاعتماد على النموذج على ميزات زائفة. الاستفادة من نطق رواية من CAD لتحليل علاقتهم بأداء نموذجي، نجد أن CAD الذي يعمل على البناء مباشرة أو مجموعة متنوعة من CAD يؤدي إلى أداء أعلى.
As NLP models are increasingly deployed in socially situated settings such as online abusive content detection, it is crucial to ensure that these models are robust. One way of improving model robustness is to generate counterfactually augmented data (CAD) for training models that can better learn to distinguish between core features and data artifacts. While models trained on this type of data have shown promising out-of-domain generalizability, it is still unclear what the sources of such improvements are. We investigate the benefits of CAD for social NLP models by focusing on three social computing constructs --- sentiment, sexism, and hate speech. Assessing the performance of models trained with and without CAD across different types of datasets, we find that while models trained on CAD show lower in-domain performance, they generalize better out-of-domain. We unpack this apparent discrepancy using machine explanations and find that CAD reduces model reliance on spurious features. Leveraging a novel typology of CAD to analyze their relationship with model performance, we find that CAD which acts on the construct directly or a diverse set of CAD leads to higher performance.
References used
https://aclanthology.org/
rate research
Read More
This paper presents multidimensional Social Opinion Mining on user-generated content gathered from newswires and social networking services in three different languages: English ---a high-resourced language, Maltese ---a low-resourced language, and M
altese-English ---a code-switched language. Multiple fine-tuned neural classification language models which cater for the i) English, Maltese and Maltese-English languages as well as ii) five different social opinion dimensions, namely subjectivity, sentiment polarity, emotion, irony and sarcasm, are presented. Results per classification model for each social opinion dimension are discussed.
Natural conversations are filled with disfluencies. This study investigates if and how BERT understands disfluency with three experiments: (1) a behavioural study using a downstream task, (2) an analysis of sentence embeddings and (3) an analysis of
the attention mechanism on disfluency. The behavioural study shows that without fine-tuning on disfluent data, BERT does not suffer significant performance loss when presented disfluent compared to fluent inputs (exp1). Analysis on sentence embeddings of disfluent and fluent sentence pairs reveals that the deeper the layer, the more similar their representation (exp2). This indicates that deep layers of BERT become relatively invariant to disfluency. We pinpoint attention as a potential mechanism that could explain this phenomenon (exp3). Overall, the study suggests that BERT has knowledge of disfluency structure. We emphasise the potential of using BERT to understand natural utterances without disfluency removal.
Measuring event salience is essential in the understanding of stories. This paper takes a recent unsupervised method for salience detection derived from Barthes Cardinal Functions and theories of surprise and applies it to longer narrative forms. We
improve the standard transformer language model by incorporating an external knowledgebase (derived from Retrieval Augmented Generation) and adding a memory mechanism to enhance performance on longer works. We use a novel approach to derive salience annotation using chapter-aligned summaries from the Shmoop corpus for classic literary works. Our evaluation against this data demonstrates that our salience detection model improves performance over and above a non-knowledgebase and memory augmented language model, both of which are crucial to this improvement.
There are two approaches for pairwise sentence scoring: Cross-encoders, which perform full-attention over the input pair, and Bi-encoders, which map each input independently to a dense vector space. While cross-encoders often achieve higher performan
ce, they are too slow for many practical use cases. Bi-encoders, on the other hand, require substantial training data and fine-tuning over the target task to achieve competitive performance. We present a simple yet efficient data augmentation strategy called Augmented SBERT, where we use the cross-encoder to label a larger set of input pairs to augment the training data for the bi-encoder. We show that, in this process, selecting the sentence pairs is non-trivial and crucial for the success of the method. We evaluate our approach on multiple tasks (in-domain) as well as on a domain adaptation task. Augmented SBERT achieves an improvement of up to 6 points for in-domain and of up to 37 points for domain adaptation tasks compared to the original bi-encoder performance.
Transformers-based pretrained language models achieve outstanding results in many well-known NLU benchmarks. However, while pretraining methods are very convenient, they are expensive in terms of time and resources. This calls for a study of the impa
ct of pretraining data size on the knowledge of the models. We explore this impact on the syntactic capabilities of RoBERTa, using models trained on incremental sizes of raw text data. First, we use syntactic structural probes to determine whether models pretrained on more data encode a higher amount of syntactic information. Second, we perform a targeted syntactic evaluation to analyze the impact of pretraining data size on the syntactic generalization performance of the models. Third, we compare the performance of the different models on three downstream applications: part-of-speech tagging, dependency parsing and paraphrase identification. We complement our study with an analysis of the cost-benefit trade-off of training such models. Our experiments show that while models pretrained on more data encode more syntactic knowledge and perform better on downstream applications, they do not always offer a better performance across the different syntactic phenomena and come at a higher financial and environmental cost.
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
