Zobrazeno 1 - 10
of 54 573
pro vyhledávání: '"Sanjiv"'
Autor:
Srinivas, V. (AUTHOR) vsrinivas@nic.in
Publikováno v:
Indian Journal of Public Administration. Mar2024, Vol. 70 Issue 1, p216-219. 4p.
Publikováno v:
In Journal of Nuclear Cardiology February 2021 28(1):30-33
Autor:
Bariwal, Sanjiv Kumar, Kumar, Rajesh
The phenomenon of collisional breakage in particulate processes has garnered significant interest due to its wide-ranging applications in fields such as milling, astrophysics, and disk formation. This study investigates the analysis of the pure colli
Externí odkaz:
http://arxiv.org/abs/2412.01943
Autor:
Bariwal, Sanjiv Kumar, Kumar, Rajesh
This article focuses on the finite volume method (FVM) as an instrument tool to deal with the non-linear collisional-induced breakage equation (CBE) that arises in the particulate process. Notably, we consider the non-conservative approximation of th
Externí odkaz:
http://arxiv.org/abs/2411.16925
We propose PPLqa, an easy to compute, language independent, information-theoretic metric to measure the quality of responses of generative Large Language Models (LLMs) in an unsupervised way, without requiring ground truth annotations or human superv
Externí odkaz:
http://arxiv.org/abs/2411.15320
One of the core pillars of efficient deep learning methods is architectural improvements such as the residual/skip connection, which has led to significantly better model convergence and quality. Since then the residual connection has become ubiquito
Externí odkaz:
http://arxiv.org/abs/2411.07501
The intriguing in-context learning (ICL) abilities of deep Transformer models have lately garnered significant attention. By studying in-context linear regression on unimodal Gaussian data, recent empirical and theoretical works have argued that ICL
Externí odkaz:
http://arxiv.org/abs/2410.21698
Autor:
Yen, Jui-Nan, Si, Si, Meng, Zhao, Yu, Felix, Duvvuri, Sai Surya, Dhillon, Inderjit S., Hsieh, Cho-Jui, Kumar, Sanjiv
Low-rank adaption (LoRA) is a widely used parameter-efficient finetuning method for LLM that reduces memory requirements. However, current LoRA optimizers lack transformation invariance, meaning the actual updates to the weights depends on how the tw
Externí odkaz:
http://arxiv.org/abs/2410.20625
Autor:
Rawat, Ankit Singh, Sadhanala, Veeranjaneyulu, Rostamizadeh, Afshin, Chakrabarti, Ayan, Jitkrittum, Wittawat, Feinberg, Vladimir, Kim, Seungyeon, Harutyunyan, Hrayr, Saunshi, Nikunj, Nado, Zachary, Shivanna, Rakesh, Reddi, Sashank J., Menon, Aditya Krishna, Anil, Rohan, Kumar, Sanjiv
A primary challenge in large language model (LLM) development is their onerous pre-training cost. Typically, such pre-training involves optimizing a self-supervised objective (such as next-token prediction) over a large corpus. This paper explores a
Externí odkaz:
http://arxiv.org/abs/2410.18779
We present a novel soft prompt based framework, SoftSRV, that leverages a frozen pre-trained large language model (LLM) to generate targeted synthetic text sequences. Given a sample from the target distribution, our proposed framework uses data-drive
Externí odkaz:
http://arxiv.org/abs/2410.16534