Text Classification
Text classification aims to automatically categorize text into predefined categories, driven by the need for efficient and accurate information processing across diverse domains. Current research focuses on leveraging large language models (LLMs) like BERT and Llama 2, often enhanced with techniques such as fine-tuning, data augmentation, and active learning, alongside traditional machine learning methods like SVMs and XGBoost. These advancements are improving the accuracy and efficiency of text classification, with significant implications for applications ranging from medical diagnosis and financial analysis to social media monitoring and legal research. A key challenge remains ensuring model robustness, interpretability, and fairness, particularly when dealing with imbalanced datasets or noisy labels.
Papers
Novel Interpretable and Robust Web-based AI Platform for Phishing Email Detection
Abdulla Al-Subaiey, Mohammed Al-Thani, Naser Abdullah Alam, Kaniz Fatema Antora, Amith Khandakar, SM Ashfaq Uz Zaman
Simple-Sampling and Hard-Mixup with Prototypes to Rebalance Contrastive Learning for Text Classification
Mengyu Li, Yonghao Liu, Fausto Giunchiglia, Xiaoyue Feng, Renchu Guan
On Adversarial Examples for Text Classification by Perturbing Latent Representations
Korn Sooksatra, Bikram Khanal, Pablo Rivas
Liberating Seen Classes: Boosting Few-Shot and Zero-Shot Text Classification via Anchor Generation and Classification Reframing
Han Liu, Siyang Zhao, Xiaotong Zhang, Feng Zhang, Wei Wang, Fenglong Ma, Hongyang Chen, Hong Yu, Xianchao Zhang