arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Jayashree

1,406 papers · 1877–2026

1967: 01968: 01969: 01970: 01971: 01972: 01973: 01974: 01975: 11976: 11977: 01978: 01979: 21980: 01981: 01982: 01983: 01984: 01985: 11986: 31987: 01988: 01989: 11990: 01991: 01992: 21993: 01994: 21995: 21996: 41997: 31998: 51999: 62000: 32001: 22002: 92003: 52004: 32005: 62006: 142007: 102008: 142009: 272010: 252011: 262012: 372013: 362014: 612015: 882016: 912017: 802018: 532019: 872020: 792021: 922022: 932023: 1302024: 1202025: 1182026: 62
1967peak 130 in 20232026

Frequent co-authors

Papers

showing 1–25 of 1,406 RSS

Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve

Amey Agrawal, Nitin Kedia, Ashish Panwar +5 more

cs.LG · · cited 874× · arXiv ↗ · more like this ·

Each LLM serving request goes through two phases. The first is prefill which processes the entire input prompt and produces the first output token and the second is decode which generates the rest of output tokens, one-at-a-time. Prefill iterations have high latency but saturate GPU compute due to parallel processing of the input prompt. In contrast, decode iterations have low latency but also low compute utilization because a decode iteration processes only a single token per request. This makes batching highly effective for decodes and consequently for overall throughput. However, batching multiple requests leads to an interleaving of prefill and decode iterations which makes it challenging to achieve both high throughput and low latency. We introduce an efficient LLM inference scheduler, Sarathi-Serve, to address this throughput-latency tradeoff. Sarathi-Serve introduces…

SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills

Amey Agrawal, Ashish Panwar, Jayashree Mohan +3 more

cs.LG · · cited 291× · arXiv ↗ · more like this ·

Large Language Model (LLM) inference consists of two distinct phases - prefill phase which processes the input prompt and decode phase which generates output tokens autoregressively. While the prefill phase effectively saturates GPU compute at small batch sizes, the decode phase results in low compute utilization as it generates one token at a time per request. The varying prefill and decode times also lead to imbalance across micro-batches when using pipeline parallelism, resulting in further inefficiency due to bubbles. We present SARATHI to address these challenges. SARATHI employs chunked-prefills, which splits a prefill request into equal sized chunks, and decode-maximal batching, which constructs a batch using a single prefill chunk and populates the remaining slots with decodes. During inference, the prefill chunk saturates GPU compute, while the decode requests 'piggyback' and…

Towards Pose Invariant Face Recognition in the Wild

Jian Zhao, Cheng Yu, Yan Xu +9 more

Computer Vision and Pattern Recognition · · cited 273× · DOI ↗ · more like this ·

Pose variation is one key challenge in face recognition. As opposed to current techniques for pose invariant face recognition, which either directly extract pose invariant features for recognition, or first normalize profile face images to frontal pose before feature extraction, we argue that it is more desirable to perform both tasks jointly to allow them to benefit from each other. To this end, we propose a Pose Invariant Model (PIM) for face recognition in the wild, with three distinct novelties. First, PIM is a novel and unified deep architecture, containing a Face Frontalization sub-Net (FFN) and a Discriminative Learning sub-Net (DLN), which are jointly learned from end to end. Second, FFN is a well-designed dual-path Generative Adversarial Network (GAN) which simultaneously perceives global structures and local details, incorporated with an unsupervised cross-domain adversarial…

Attracting tourists to travel companies’ websites: the structural relationship between website brand, personal value, shopping experience, perceived risk and purchase intention

Shahriar Mohseni -, Sreenivasan Jayashree, Sajad Rezaei +2 more

Sociology and Political Science · Current Issues in Tourism · · cited 198× · DOI ↗ · more like this ·

This article aims to examine the structural relationship between website brand, personal value, shopping experience, perceived risk and purchase intention from travel websites. Built upon the theory of consumers’ perceived risk, a theoretical model was proposed and a questionnaire was developed. The fieldwork utilized responses from 409 participants who purchased travel items from websites of Malaysian travel companies. Partial least square (PLS) path modelling approach, a variance-based structural equation modelling (VB-SEM), was used to assess the overall goodness-of-fit tests, measurement and structural model. The results highlight different aspects related to the effectiveness and attractiveness of travel companies’ websites. Its unique finding highlights the importance of personal value as a user characteristic factor that can strongly affect online purchase intention. In addition…

RECIPE : Converting Concurrent DRAM Indexes to Persistent-Memory Indexes

Se Kwon Lee, Jayashree Mohan, Sanidhya Kashyap +2 more

cs.DC · · cited 189× · arXiv ↗ · more like this ·

We present Recipe, a principled approach for converting concurrent DRAM indexes into crash-consistent indexes for persistent memory (PM). The main insight behind Recipe is that isolation provided by a certain class of concurrent in-memory indexes can be translated with small changes to crash-consistency when the same index is used in PM. We present a set of conditions that enable the identification of this class of DRAM indexes, and the actions to be taken to convert each index to be persistent. Based on these conditions and conversion actions, we modify five different DRAM indexes based on B+ trees, tries, radix trees, and hash tables to their crash-consistent PM counterparts. The effort involved in this conversion is minimal, requiring 30-200 lines of code. We evaluated the converted PM indexes on Intel DC Persistent Memory, and found that they outperform state-of-the-art…

Video-based Person Re-identification with Accumulative Motion Context

Hao Liu, Zequn Jie, Karlekar Jayashree +4 more

cs.CV · · cited 181× · arXiv ↗ · more like this ·

Video based person re-identification plays a central role in realistic security and video surveillance. In this paper we propose a novel Accumulative Motion Context (AMOC) network for addressing this important problem, which effectively exploits the long-range motion context for robustly identifying the same person under challenging conditions. Given a video sequence of the same or different persons, the proposed AMOC network jointly learns appearance representation and motion context from a collection of adjacent frames using a two-stream convolutional architecture. Then AMOC accumulates clues from motion context by recurrent aggregation, allowing effective information flow among adjacent frames and capturing dynamic gist of the persons. The architecture of AMOC is end-to-end trainable and thus motion context can be adapted to complement appearance clues under unfavorable conditions…

Split Learning for collaborative deep learning in healthcare

Maarten G. Poirot, Praneeth Vepakomma, Ken Chang +3 more

cs.LG · · cited 180× · arXiv ↗ · more like this ·

Shortage of labeled data has been holding the surge of deep learning in healthcare back, as sample sizes are often small, patient information cannot be shared openly, and multi-center collaborative studies are a burden to set up. Distributed machine learning methods promise to mitigate these problems. We argue for a split learning based approach and apply this distributed learning method for the first time in the medical field to compare performance against (1) centrally hosted and (2) non collaborative configurations for a range of participants. Two medical deep learning tasks are used to compare split learning to conventional single and multi center approaches: a binary classification problem of a data set of 9000 fundus photos, and multi-label classification problem of a data set of 156,535 chest X-rays. The several distributed learning setups are compared for a range of 1-50…

Analyzing and Mitigating Data Stalls in DNN Training

Jayashree Mohan, Amar Phanishayee, Ashish Raniwala +1 more

cs.DC · · cited 166× · arXiv ↗ · more like this ·

Training Deep Neural Networks (DNNs) is resource-intensive and time-consuming. While prior research has explored many different ways of reducing DNN training time, the impact of input data pipeline, i.e., fetching raw data items from storage and performing data pre-processing in memory, has been relatively unexplored. This paper makes the following contributions: (1) We present the first comprehensive analysis of how the input data pipeline affects the training time of widely-used computer vision and audio Deep Neural Networks (DNNs), that typically involve complex data preprocessing. We analyze nine different models across three tasks and four datasets while varying factors such as the amount of memory, number of CPU threads, storage device, GPU generation etc on servers that are a part of a large production cluster at Microsoft. We find that in many cases, DNN training time is…

Neural Person Search Machines

Hao Liu, Jiashi Feng, Zequn Jie +5 more

cs.CV · · cited 159× · arXiv ↗ · more like this ·

We investigate the problem of person search in the wild in this work. Instead of comparing the query against all candidate regions generated in a query-blind manner, we propose to recursively shrink the search area from the whole image till achieving precise localization of the target person, by fully exploiting information from the query and contextual cues in every recursive search step. We develop the Neural Person Search Machines (NPSM) to implement such recursive localization for person search. Benefiting from its neural search mechanism, NPSM is able to selectively shrink its focus from a loose region to a tighter one containing the target automatically. In this process, NPSM employs an internal primitive memory component to memorize the query representation which modulates the attention and augments its robustness to other distracting regions. Evaluations on two benchmark…

vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention

Ramya Prabhu, Ajay Nayak, Jayashree Mohan +2 more

cs.LG · · cited 156× · arXiv ↗ · more like this ·

PagedAttention is a popular approach for dynamic memory allocation in LLM serving systems. It enables on-demand allocation of GPU memory to mitigate KV cache fragmentation -- a phenomenon that crippled the batch size (and consequently throughput) in prior systems. However, in trying to allocate physical memory at runtime, PagedAttention ends up changing the virtual memory layout of the KV cache from contiguous to non-contiguous. Such a design leads to non-trivial programming and performance overheads. We present vAttention -- an approach that mitigates fragmentation in physical memory while retaining the contiguity of KV cache in virtual memory. We achieve this by decoupling the allocation of virtual and physical memory using CUDA virtual memory management APIs. We also introduce various LLM-specific optimizations to address the limitations of CUDA virtual memory support. Overall…

Machine Learning in Automatic Speech Recognition: A Survey

Jayashree Padmanabhan, Melvin Jose Johnson Premkumar

Artificial Intelligence · IETE Technical Review · · cited 151× · DOI ↗ · more like this ·

Over the past few decades, there has been tremendous development in machine learning paradigms used in automatic speech recognition (ASR) for home automation to space exploration. Though commercial speech recognizers are available for certain well-defined applications like dictation and transcription, many issues in ASR like recognition in noisy environments, multilingual recognition, and multi-modal recognition are yet to be addressed effectively. A comprehensive review of common machine learning techniques like artificial neural networks, support vector machines, and Gaussian mixture models along with hidden Markov models employed in ASR is provided. A thorough review on the recent developments in deep learning which has provided significant improvements in ASR performance, along with its relevance in the future of ASR, is also presented.

Nomophobia: an emerging issue in medical institutions?

Iqbal Ahmad Farooqui, Prasad Pore, Jayashree Sachin Gothankar

Sociology and Political Science · Journal of Mental Health · · cited 134× · DOI ↗ · more like this ·

BACKGROUND: Nomophobia is No Mobile Phone Phobia - the fear of being without a mobile device or beyond mobile phone contact. The time and money wastage by medical students is a negative aspect of mobile phone dependence. This time can be utilized for academic and recreational activities. AIM: To assess the prevalence of Nomophobia in the students in 1st year MBBS. METHODS: Study area: A medical college in Pune city. STUDY PERIOD: December 2015 to February 2016. A cross-sectional study was carried out on all the students of 1st year MBBS. A predesigned and pretested questionnaire was used to collect data. Data were analyzed statistically by simple proportions using SPSS v20. RESULTS: A total of 145 students were monitored according to inclusion and exclusion criteria. Amongst all the participants, 45.5% were males (66/145) and 54.5% were females (79/145). Mild Nomophobia was found in…

Dual-Agent GANs for Photorealistic and Identity Preserving Profile Face Synthesis

Jian Zhao, Lin Xiong, Panasonic Karlekar Jayashree +7 more

Computer Vision and Pattern Recognition · Neural Information Processing Systems · · cited 129× · OpenAlex ↗ · more like this ·

Synthesizing realistic profile faces is promising for more efficiently training deep pose-invariant models for large-scale unconstrained face recognition, by populating samples with extreme poses and avoiding tedious annotations. However, learning from synthetic faces may not achieve the desired performance due to the discrepancy between distributions of the synthetic and real face images. To narrow this gap, we propose a Dual-Agent Generative Adversarial Network (DA-GAN) model, which can improve the realism of a face simulator's output using unlabeled real faces, while preserving the identity information during the realism refinement. The dual agents are specifically designed for distinguishing real v.s. fake and identities simultaneously. In particular, we employ an off-the-shelf 3D face model as a simulator to generate profile face images with varying poses. DA-GAN leverages a fully…

Online impulse buying of tourism products

Sajad Rezaei, Faizan Ali, Muslim Amin +1 more

Sociology and Political Science · Journal of Hospitality and Tourism Technology · · cited 128× · DOI ↗ · more like this ·

Purpose – The purpose of this paper is to examine the structural relationship between web site personality, utilitarian web browsing, hedonic web browsing and online impulse buying of tourism products. Design/methodology/approach – A total of 405 valid online questionnaires were collected to empirically test the measurement and structural model using partial least square path modelling approach, a variance-based structural equation modelling technique. The study sample includes experienced online shoppers who performed shopping tourism products and services via internet medium. Findings – The results imply that web site personality is a second-order reflective construct comprising solidity, enthusiasm, genuineness, sophistication and unpleasantness. web site personality positively influences utilitarian web browsing, hedonic web browsing and online impulse buying; and both hedonic web…

Assessing the (Un)Trustworthiness of Saliency Maps for Localizing Abnormalities in Medical Imaging

Nishanth Arun, Nathan Gaw, Praveer Singh +10 more

cs.CV · · cited 115× · arXiv ↗ · more like this ·

Saliency maps have become a widely used method to make deep learning models more interpretable by providing post-hoc explanations of classifiers through identification of the most pertinent areas of the input medical image. They are increasingly being used in medical imaging to provide clinically plausible explanations for the decisions the neural network makes. However, the utility and robustness of these visualization maps has not yet been rigorously examined in the context of medical imaging. We posit that trustworthiness in this context requires 1) localization utility, 2) sensitivity to model weight randomization, 3) repeatability, and 4) reproducibility. Using the localization information available in two large public radiology datasets, we quantify the performance of eight commonly used saliency map approaches for the above criteria using area under the precision-recall curves…

Fair Conformal Predictors for Applications in Medical Imaging

Charles Lu, Andreanne Lemay, Ken Chang +2 more

eess.IV · · cited 109× · arXiv ↗ · more like this ·

Deep learning has the potential to automate many clinically useful tasks in medical imaging. However translation of deep learning into clinical practice has been hindered by issues such as lack of the transparency and interpretability in these "black box" algorithms compared to traditional statistical methods. Specifically, many clinical deep learning models lack rigorous and robust techniques for conveying certainty (or lack thereof) in their predictions -- ultimately limiting their appeal for extensive use in medical decision-making. Furthermore, numerous demonstrations of algorithmic bias have increased hesitancy towards deployment of deep learning for clinical applications. To this end, we explore how conformal predictions can complement existing deep learning approaches by providing an intuitive way of expressing uncertainty while facilitating greater transparency to clinical…

Overview of the ImageCLEF 2012 medical image retrieval and classiFIcation tasks

Henning Müller, Alba García Seco de Herrera, Jayashree Kalpathy–Cramer +3 more

Computer Vision and Pattern Recognition · · cited 108× · OpenAlex ↗ · more like this ·

The ninth edition of the ImageCLEF medical image retrieval and classication tasks was organized in 2012. A subset of the open access collection of PubMed Central was used as the database in 2012, using a larger number of over 300'000 images than in 2011. As in previous years, there were three subtasks: modality classication, image{based and case{based retrieval. A new hierarchy for article gures was created for the modality classi- cation task. The modality detection could be one of the most important lters to limit the search and focus the results sets. The goal of the image{based and the case{based retrieval tasks were similar compared to 2011 adding mainly complexity. The number of groups submitting runs has remained stable at 17, with the number of submitted runs remaining roughly the same with 202 (207 in 2011). Of these, 122 were image{based retrieval runs, 37 were case{based runs…

Overview of the ImageCLEF 2013 medical tasks

Alba García Seco de Herrera, Jayashree Kalpathy–Cramer, Dina Demner‐Fushman +2 more

Computer Vision and Pattern Recognition · · cited 101× · OpenAlex ↗ · more like this ·

In 2013, the tenth edition of the medical task of the Image- CLEF benchmark was organized. For the rst time, the ImageCLEFmed workshop takes place in the United States of America at the annual AMIA (American Medical Informatics Association) meeting even though the task was organized as in previous years in connection with the other ImageCLEF tasks. Like 2012, a subset of the open access collection of PubMed Central was distributed. This year, there were four subtasks: modality classication, compound gure separation, image{based and case{based retrieval. The compound gure separation task was included due to the large number of multipanel images available in the literature and the importance to separate them for targeted retrieval. More com- pound gures were also included in the modality classication task to make it correspond to the distribution in the full database. The retrieval tasks…

Real-Time Static Hand Gesture Recognition for American Sign Language (ASL) in Complex Background

Jayashree R. Pansare, S. H. Gawande, Maya Ingle

Human-Computer Interaction · Journal of Signal and Information Processing · · cited 101× · DOI ↗ · more like this ·

Hand gestures are powerful means of communication among humans and sign language is the most natural and expressive way of communication for dump and deaf people. In this work, real-time hand gesture system is proposed. Experimental setup of the system uses fixed position low-cost web camera with 10 mega pixel resolution mounted on the top of monitor of computer which captures snapshot using Red Green Blue [RGB] color space from fixed distance. This work is divided into four stages such as image preprocessing, region extraction, feature extraction, feature matching. First stage converts captured RGB image into binary image using gray threshold method with noise removed using median filter [medfilt2] and Guassian filter, followed by morphological operations. Second stage extracts hand region using blob and crop is applied for getting region of interest and then “Sobel” edge detection is…

Modelling the significance of social media marketing activities, brand equity and loyalty to predict consumers’ willingness to pay premium price for portable tech gadgets

Chinnasamy Agamudainambhi Malarvizhi, Abdullah Al Mamun, Sreenivasan Jayashree +2 more

Sociology and Political Science · Heliyon · · cited 98× · DOI ↗ · more like this ·

In order to sustain business operations during the COVID-19 pandemic, nearly all industries have to adopt online technology and social media marketing activities (SMMAs). Globally, portable tech gadgets are rapidly expanding, but empirical studies on SMMAs in relation to portable tech gadgets in Malaysia have remained scarce. Therefore, this study examined the elements of SMMAs and their influence on brand equity in terms of brand awareness (BBA) and brand image (BBI) as well as brand loyalty (BRL) and willingness to pay premium price (WPP) among Malaysian consumers of portable tech gadgets users. Five components of SMMAs, namely entertainment (ENT), interactivity (INT), trendiness (TRE), customisation (CUS), and electronic word-of-mouth (EWOM), were examined to understand how SMMAs influence BBA, BBI, BRL, and WPP. An online survey was conducted with 1332 Malaysian youths who used…

Finding Crash-Consistency Bugs with Bounded Black-Box Crash Testing

Jayashree Mohan, Ashlie Martinez, Soujanya Ponnapalli +2 more

cs.OS · · cited 98× · arXiv ↗ · more like this ·

We present a new approach to testing file-system crash consistency: bounded black-box crash testing (B3). B3 tests the file system in a black-box manner using workloads of file-system operations. Since the space of possible workloads is infinite, B3 bounds this space based on parameters such as the number of file-system operations or which operations to include, and exhaustively generates workloads within this bounded space. Each workload is tested on the target file system by simulating power-loss crashes while the workload is being executed, and checking if the file system recovers to a correct state after each crash. B3 builds upon insights derived from our study of crash-consistency bugs reported in Linux file systems in the last five years. We observed that most reported bugs can be reproduced using small workloads of three or fewer file-system operations on a newly-created file…

Visual and Spatial Modes in Science Learning

Jayashree Ramadas

Education · International Journal of Science Education · · cited 97× · DOI ↗ · more like this ·

Abstract This paper surveys some major trends from research on visual and spatial thinking coming from cognitive science, developmental psychology, science literacy, and science studies. It explores the role of visualisation in creativity, in building mental models, and in the communication of scientific ideas, in order to place these findings in the context of science education research and practice. Keywords: Science educationTransformational reasoningVisual and spatial learning Notes The original version of this paper is a part of epiSTEME‐2 reviews, edited by B. Choksi and C. Natarajan and published in 2008 by Macmillan India Ltd. Visual presentation at epiSTEME‐2 held at the Homi Bhabha Centre for Science Education, TIFR, Mumbai, in February 2007. See http://www.hbcse.tifr.res.in/ep2visual-presentation.pdf.

Family Cultural Socialization, Ethnic Identity, and Self-Esteem

Jayashree Mohanty, Gary Keokse, Esther Sales

Safety Research · Journal of Ethnic & Cultural Diversity in Social Work · · cited 96× · DOI ↗ · more like this ·

Abstract Using a Web-based survey, this study of 82 adult international adoptees examined the relationship between parental support for cultural socialization and its effect on adoptees' selfesteem. Feelings of belongingness and ethnic identity were predicted to serve as mediators between the central variables. The results showed a positive relationship between cultural socialization and selfesteem, which was mediated by a feeling of belongingness and one aspect of ethnic identification (marginality) among Asianborn international adoptees. These data suggest that counselors should raise awareness and knowledge of adoptive parents about the importance of cultural continuity in the child's upbringing.

Conditional Convolutional Neural Network for Modality-Aware Face Recognition

Chao Xiong, Xiaowei Zhao, Danhang Tang +3 more

Computer Vision and Pattern Recognition · · cited 92× · DOI ↗ · more like this ·

Faces in the wild are usually captured with various poses, illuminations and occlusions, and thus inherently multimodally distributed in many tasks. We propose a conditional Convolutional Neural Network, named as c-CNN, to handle multimodal face recognition. Different from traditional CNN that adopts fixed convolution kernels, samples in c-CNN are processed with dynamically activated sets of kernels. In particular, convolution kernels within each layer are only sparsely activated when a sample is passed through the network. For a given sample, the activations of convolution kernels in a certain layer are conditioned on its present intermediate representation and the activation status in the lower layers. The activated kernels across layers define the sample-specific adaptive routes that reveal the distribution of underlying modalities. Consequently, the proposed framework does not rely…

Finding faces in photographs

A. N. Rajagopalan, K. Sunil Kumar, Jayashree Karlekar +5 more

Computer Vision and Pattern Recognition · DSpace (IIT Bombay) · · cited 92× · DOI ↗ · more like this ·

Two new schemes are presented for finding human faces in a photograph. The first scheme approximates the unknown distributions of the face and the face-like manifolds wing higher order statistics (HOS). An HOS-based data clustering algorithm is also proposed. In the second scheme, the face to non-face and non-face to face transitions are learnt using a hidden Markov model (HMM). The HMM parameters are estimated corresponding to a given photograph and the faces are located by examining the optimal state sequence of the HMM. Experimental results are presented on the performance of both the schemes.