Výsledky vyhledávání - "Hassanpour, Mohammad"

Report

Single Parent Family: A Spectrum of Family Members from a Single Pre-Trained Foundation Model

Autor: Hajimolahoseini, Habib, Hassanpour, Mohammad, Ataiefard, Foozhan, Chen, Boxing, Liu, Yang

This paper introduces a novel method of Progressive Low Rank Decomposition (PLRD) tailored for the compression of large language models. Our approach leverages a pre-trained model, which is then incrementally decompressed to smaller sizes using progr

Externí odkaz: http://arxiv.org/abs/2406.19995

Zobrazit plný text záznamu

Report

SkipViT: Speeding Up Vision Transformers with a Token-Level Skip Connection

Autor: Ataiefard, Foozhan, Ahmed, Walid, Hajimolahoseini, Habib, Asani, Saina, Javadi, Farnoosh, Hassanpour, Mohammad, Awad, Omar Mohamed, Wen, Austin, Liu, Kangling, Liu, Yang

Vision transformers are known to be more computationally and data-intensive than CNN models. These transformer models such as ViT, require all the input image tokens to learn the relationship among them. However, many of these tokens are not informat

Externí odkaz: http://arxiv.org/abs/2401.15293

Zobrazit plný text záznamu

Report

Test-Time Personalization with Meta Prompt for Gaze Estimation

Autor: Liu, Huan, Qi, Julia, Li, Zhenhao, Hassanpour, Mohammad, Wang, Yang, Plataniotis, Konstantinos, Yu, Yuanhao

Despite the recent remarkable achievement in gaze estimation, efficient and accurate personalization of gaze estimation without labels is a practical problem but rarely touched on in the literature. To achieve efficient personalization, we take inspi

Externí odkaz: http://arxiv.org/abs/2401.01577

Zobrazit plný text záznamu

Report

Do LLMs Work on Charts? Designing Few-Shot Prompts for Chart Question Answering and Summarization

Autor: Do, Xuan Long, Hassanpour, Mohammad, Masry, Ahmed, Kavehzadeh, Parsa, Hoque, Enamul, Joty, Shafiq

A number of tasks have been proposed recently to facilitate easy access to charts such as chart QA and summarization. The dominant paradigm to solve these tasks has been to fine-tune a pretrained model on the task data. However, this approach is not

Externí odkaz: http://arxiv.org/abs/2312.10610

Zobrazit plný text záznamu

Report

GQKVA: Efficient Pre-training of Transformers by Grouping Queries, Keys, and Values

Autor: Javadi, Farnoosh, Ahmed, Walid, Hajimolahoseini, Habib, Ataiefard, Foozhan, Hassanpour, Mohammad, Asani, Saina, Wen, Austin, Awad, Omar Mohamed, Liu, Kangling, Liu, Yang

Massive transformer-based models face several challenges, including slow and computationally intensive pre-training and over-parametrization. This paper addresses these challenges by proposing a versatile method called GQKVA, which generalizes query,

Externí odkaz: http://arxiv.org/abs/2311.03426

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání