Scholar
Ashim Dahal
Google Scholar ID: Nt9K4nsAAAAJ
Research Assistant, University of Southern Mississippi
Citations & Impact
Citations
9
H-index
2
i10-index
0
Publications
12
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- 1. POVQA: Preference-Optimized Video Question Answering with Rationales for Data Efficiency (arXiv, 2025)
- 2. Redemption Score: A Multi-Modal Evaluation Framework for Image Captioning via Distributional, Perceptual, and Linguistic Signal Triangulation (arXiv, 2025)
- 3. EEG-to-Text Translation: A Model for Deciphering Human Brain Activity (arXiv, 2025)
- 4. Embedding Shift Dissection on CLIP: Effects of Augmentations on VLM's Representation Learning (CVPRW, 2025)
- 5. Heuristical Comparison of Vision Transformers Against Convolutional Neural Networks for Semantic Segmentation on Remote Sensing Imagery (IEEE Sensors Journal, 2025)
- 6. Multi-Lingual Cyber Threat Detection in Tweets/X Using ML, DL, and LLM: A Comparative Analysis (IEEE Transactions on Computational Social Systems, 2025)
- 7. Efficiency Bottlenecks of Convolutional Kolmogorov-Arnold Networks: A Comprehensive Scrutiny with ImageNet, AlexNet, LeNet and Tabular Classification (arXiv, 2025)
- 8. Awarded $5500 summer research grant in 2025 by the Drapeau Center for Undergraduate Research (DCUR) for research on Gaussian Splatting.
Research Experience
- Works in the Cyber Innovations Lab, focusing on modern deep learning techniques in computer vision and multimodal systems. Closely collaborates with Dr. Nick Rahimi and his graduate and undergraduate students on topics such as Stable Diffusion, Vision Transformers, Kolmogorov Arnold Networks, cybersecurity, visual question answering, NLP, and brain-computer interface.
Education
- Undergraduate Research Assistant at the University of Southern Mississippi, advised by Dr. Nick Rahimi.
Background
- Research interests include computer vision, deep learning, generative AI, and image processing. Focused on understanding scenes, analyzing deep learning techniques, and inferring meaningful information by tweaking model architectures and their loss function optimization.
Miscellany
- 1. Lead Organizer of the Google Developers Group (GDG) On Campus at USM
- 2. Former Head of Artificial Intelligence at Google Developers Student Club at USM
- 3. Research Liaison representing the School of Computing Science and Computer Engineering
- 4. Hobbies include playing bansuri and reading literature
- 5. Won the Eagles Write Award (best assignment among all freshmen) 2023/2024 for an essay written in 45 minutes
