A5052035014- Topic Modeling
- Natural Language Processing Techniques
- Text and Document Classification Technologies
- Speech Recognition and Synthesis
- Multimodal Machine Learning Applications
- Advanced Text Analysis Techniques
- Reproductive Biology and Fertility
- Seismology and Earthquake Studies
- Domain Adaptation and Few-Shot Learning
- Expert finding and Q&A systems
- Adversarial Robustness in Machine Learning
- Sperm and Testicular Function
University of Guilan
2018-2024
Sharif University of Technology
2023
Sperm Morphology Analysis (SMA) is pivotal in diagnosing male infertility. However, manual analysis subjective and time-intensive. Artificial intelligence presents automated alternatives, but hurdles like limited data image quality constraints hinder its efficacy. These challenges impede Deep Learning (DL) models from grasping crucial sperm features. A solution enabling DL to learn sample nuances, even with data, would be invaluable. This study proposes a Knowledge Distillation (KD) method...
Transformer-based models have made remarkable advancements in various NLP areas. Nevertheless, these often exhibit vulnerabilities when confronted with adversarial attacks. In this paper, we explore the effect of quantization on robustness models. Quantization usually involves mapping a high-precision real number to lower-precision value, aiming at reducing size model hand. To best our knowledge, work is first application experiments, evaluate impact BERT and DistilBERT text classification...
Yasaman Boreshban, Seyed Morteza Mirbostani, Seyedeh Fatemeh Ahmadi, Gita Shojaee, Kamani, Gholamreza Ghassem-Sani, Abolghasem Mirroshandel. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. 2023.
Seyed Morteza Mirbostani, Yasaman Boreshban, Salam Khalifa, SeyedAbolghasem Mirroshandel, Owen Rambow. Proceedings of the 61st Annual Meeting Association for Computational Linguistics (Volume 2: Short Papers). 2023.
Contemporary question answering (QA) systems, including transformer-based architectures, suffer from increasing computational and model complexity which render them inefficient for real-world applications with limited resources. Further, training or even fine-tuning such models requires a vast amount of labeled data is often not available the task at hand. In this manuscript, we conduct comprehensive analysis mentioned challenges introduce suitable countermeasures. We propose novel knowledge...