11 papers
AgentR A Stateful and Recovery-Aware Software Architecture for LLM-based Auditable Workflows
Riya Samanta, Bidyut Saha, Soumya Kanti Ghosh +1
Modern LLM-based applications increasingly require multi- stage execution, persistent intermediate state, retry seman- tics, and auditable usage accounting. However, many LLM appli…
Synergizing a Decentralized Framework with LLM-Assisted Skill and Willingness-Aware Task Assignment for Volunteer Crowdsourcing
Riya Samanta, Rituparna Bhattyacharya
Volunteer crowdsourcing or VCS platforms increasingly support education, healthcare, disaster response, and smart city applications, yet assigning volunteers to complex tasks remai…
Affordable Precision Agriculture: A Deployment-Oriented Review of Low-Cost, Low-Power Edge AI and TinyML for Resource-Constrained Farming Systems
Riya Samanta, Bidyut Saha
Precision agriculture increasingly integrates artificial intelligence to enhance crop monitoring, irrigation management, and resource efficiency. Nevertheless, the vast majority of…
Rethinking Temporal Models for TinyML: LSTM versus 1D-CNN in Resource-Constrained Devices
Bidyut Saha, Riya Samanta
Time series classification underpins applications such as human activity recognition, healthcare monitoring, and gesture detection in the IoT domain. Tiny Machine Learning enables…
Structurally Valid Log Generation using FSM-GFlowNets
Riya Samanta
Generating structurally valid and behaviorally diverse synthetic event logs for interaction-aware models is a challenging yet crucial problem, particularly in settings with limited…
Optimizing TinyML: The Impact of Reduced Data Acquisition Rates for Time Series Classification on Microcontrollers
Riya Samanta, Bidyut Saha, Soumya K. Ghosh +1
Tiny Machine Learning (TinyML) enables efficient, lowcost, and privacy preserving machine learning inference directly on microcontroller units (MCUs) connected to sensors. Optimizi…
CTG-KrEW: Generating Synthetic Structured Contextually Correlated Content by Conditional Tabular GAN with K-Means Clustering and Efficient Word Embedding
Riya Samanta, Bidyut Saha, Soumya K. Ghosh +1
Conditional Tabular Generative Adversarial Networks (CTGAN) and their various derivatives are attractive for their ability to efficiently and flexibly create synthetic tabular data…
Sustainable Volunteer Engagement: Ensuring Potential Retention and Skill Diversity for Balanced Workforce Composition in Crowdsourcing Paradigm
Riya Samanta, Soumya K Ghosh
Crowdsourcing (CS) faces the challenge of managing complex, skill-demanding tasks, which requires effective task assignment and retention strategies to sustain a balanced workforce…
TinyTNAS: GPU-Free, Time-Bound, Hardware-Aware Neural Architecture Search for TinyML Time Series Classification
Bidyut Saha, Riya Samanta, Soumya K. Ghosh +1
In this work, we present TinyTNAS, a novel hardware-aware multi-objective Neural Architecture Search (NAS) tool specifically designed for TinyML time series classification. Unlike…
Towards Sustainable Personalized On-Device Human Activity Recognition with TinyML and Cloud-Enabled Auto Deployment
Bidyut Saha, Riya Samanta, Soumya K Ghosh +1
Human activity recognition (HAR) holds immense potential for transforming health and fitness monitoring, yet challenges persist in achieving personalized outcomes and sustainabilit…
Empowering Volunteer Crowdsourcing Services: A Serverless-assisted, Skill and Willingness Aware Task Assignment Approach for Amicable Volunteer Involvement
Riya Samanta, Biswajeet Sethi, Soumya K Ghosh
Volunteer crowdsourcing (VCS) leverages citizen interaction to address challenges by utilizing individuals' knowledge and skills. Complex social tasks often require collaboration a…