30 citations · 91 across the 44 of their papers we have counts for
6 papers · 1 filter
Subspace Networks: Scaling Decentralized Training with Communication-Efficient Model Parallelism
Sameera Ramasinghe, Thalaiyasingam Ajanthan, Gil Avraham +2
Scaling models has led to significant advancements in deep learning, but training these models in decentralized settings remains challenging due to communication bottlenecks. While…
Nesterov Method for Asynchronous Pipeline Parallel Optimization
Thalaiyasingam Ajanthan, Sameera Ramasinghe, Yan Zuo +2
Pipeline Parallelism (PP) enables large neural network training on small, interconnected devices by splitting the model into multiple stages. To maximize pipeline utilization, asyn…
VI3NR: Variance Informed Initialization for Implicit Neural Representations
Chamin Hewa Koneputugodage, Yizhak Ben-Shabat, Sameera Ramasinghe +1
Implicit Neural Representations (INRs) are a versatile and powerful tool for encoding various forms of data, including images, videos, sound, and 3D shapes. A critical factor in th…
QID: Efficient Query-Informed ViTs in Data-Scarce Regimes for OCR-free Visual Document Understanding
Binh M. Le, Shaoyuan Xu, Jinmiao Fu +6
In Visual Document Understanding (VDU) tasks, fine-tuning a pre-trained Vision-Language Model (VLM) with new datasets often falls short in optimizing the vision encoder to identify…
Scaling Prompt Instructed Zero Shot Composed Image Retrieval with Image-Only Data
Yiqun Duan, Sameera Ramasinghe, Stephen Gould +1
Composed Image Retrieval (CIR) is the task of retrieving images matching a reference image augmented with a text, where the text describes changes to the reference image in natural…
DARB-Splatting: Generalizing Splatting with Decaying Anisotropic Radial Basis Functions
Hashiru Pramuditha, Vinasirajan Viruthshaan, Vishagar Arunan +4
Splatting-based 3D reconstruction methods have gained popularity with the advent of 3D Gaussian Splatting, efficiently synthesizing high-quality novel views. These methods commonly…