Výsledky vyhledávání

Report

Exploring Curriculum Learning for Vision-Language Tasks: A Study on Small-Scale Multimodal Training

Autor: Saha, Rohan, Fahim, Abrar, Fyshe, Alona, Murphy, Alex

For specialized domains, there is often not a wealth of data with which to train large machine learning models. In such limited data / compute settings, various methods exist aiming to $\textit{do more with less}$, such as finetuning from a pretraine

Externí odkaz: http://arxiv.org/abs/2410.15509

Zobrazit plný text záznamu

Report

It's Not a Modality Gap: Characterizing and Addressing the Contrastive Gap

Autor: Fahim, Abrar, Murphy, Alex, Fyshe, Alona

Multi-modal contrastive models such as CLIP achieve state-of-the-art performance in zero-shot classification by embedding input images and texts on a joint representational space. Recently, a modality gap has been reported in two-encoder contrastive

Externí odkaz: http://arxiv.org/abs/2405.18570

Zobrazit plný text záznamu

Akademický článek

An adaptive load shedding methodology for renewable integrated power systems

Autor: Sk Fahim Abrar, Nahid-Al Masood, Mohammad Jahangir Alam

Publikováno v: Heliyon, Vol 10, Iss 21, Pp e40043- (2024)

System stability issues regarding frequency and voltage in modern power systems are growing in importance as they incorporate more and more complex components. To ensure a sustainable, pollution-free power generation, modern power systems are designe

Externí odkaz: https://doaj.org/article/b0fc0c38454f43caa94d00969713a2f8

Zobrazit plný text záznamu

Report

Unsupervised Space Partitioning for Nearest Neighbor Search

Autor: Fahim, Abrar, Ali, Mohammed Eunus, Cheema, Muhammad Aamir

Approximate Nearest Neighbor Search (ANNS) in high dimensional spaces is crucial for many real-life applications (e.g., e-commerce, web, multimedia, etc.) dealing with an abundance of data. This paper proposes an end-to-end learning framework that co

Externí odkaz: http://arxiv.org/abs/2206.08091

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání