Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
2 results
Search Results
Item Restricted EXPLORING THE TRANSFERABILITY OF ADVERSARIAL EXAMPLES IN NATURAL LANGUAGE PROCESSING(Texas A&M University-Kingsville, 2024-06-21) Allahyani, Samah; Nijim, MaisIn recent years, there has been a growing concern about the vulnerability of machine learning models, particularly in the field of natural language processing (NLP). Many tasks in natural language processing, such as text classification, machine translation, and question answering, are at risk of adversarial attacks where maliciously crafted inputs can cause them to make incorrect predictions or classifications. Adversarial examples created on one model can also fool another model. The transferability of adversarial has also garnered significant attention as it is a crucial property for facilitating black-box attacks. In our comprehensive research, we employed an array of widely used NLP models for sentiment analysis and text classification tasks. We first generated adversarial examples for a set of source models, using five state-of-the-art attack methods. We then evaluated the transferability of these adversarial examples by testing their effectiveness on different target models, to explore the main factors such as model architecture, dataset characteristics and the perturbation techniques impacting transferability. Moreover, we extended our investigation by delving into transferability-enhancing techniques. We assisted two transferability-enhancing methods and leveraged the power of Large Language Models (LLM) to generate natural adversarial examples that show a moderate transferability across different NLP architecture. Through our research, we aim to provide insights into the transferability of adversarial examples in NLP, and shed light on the factors that contribute to their transferability. This knowledge can then be used to develop more robust, and resilient, NLP models that are less susceptible to adversarial attacks; ultimately, enhancing the security and reliability of these systems in various applications.9 0Item Restricted Enhancing Clarity and Readability in Scientific Writing: An Automated Approach to Identifying Shapeless Sentences(Saudi Digital Library, 2023-11-02) Kamal, Ayah; Lopez, AdamEffective communication is essential in academic writing, where clear and coherent writing ensures research findings are disseminated effectively. However, conveying complex concepts in a readable manner remains a challenge in scientific writing. This thesis investigates automating the application of principles from the book Style: Lessons in Clarity and Grace by Williams [32] to improve the readability of scientific writing. The research focuses on identifying “shapeless” sentences that lack structure and clarity. A dataset of scientific sentences sourced from the Elsevier OA Corpus was manually annotated as “Structured”, “Shapeless” or “N/A” based on principles from Style. A Large Language Model, LLaMA-2, was fine-tuned on this dataset to classify the sentences. Optimization techniques like QLoRA enabled efficient fine-tuning within resource constraints. While, prompt engineering and few-shot learning were used to optimize inference. The fine-tuned model achieved promising accuracy in distinguishing between “Structured” and “Shapeless” sentences. The research demonstrates potential for using fine-tuned language models to automate the application of stylistic principles and enhance scientific writing. Further work is needed to expand the dataset, refine definitions, and optimize model training. Overall, this thesis establishes a foundation for using language models to identify problematic sentences and improve readability14 0