Retrieval-Augmented Reliability-Aware Selective Inference for Visual Classification
by Pratheswaran Hariharan, Haiping Xu and Donghui Yan
Journal of Engineering Research and Sciences, Volume 5, Issue 9, Page # 1-14, 2026; DOI: 10.55708/js0509001
Abstract: Multimodal large language models (MLLMs) can generate fluent visual responses even when the underlying visual prediction is weak, ambiguous, or incorrect. This work presents a retrieval-augmented, reliability-aware selective inference method that evaluates the strength and consistency of visual evidence before a prediction is communicated through a downstream multimodal response. A pretrained ResNet-50 encoder extracts normalized… Read More
(This article belongs to the Special Issue on SP8 (Special Issue on Digital and Engineering Transformations in Science and Technology (SI-DETST-26)))
A Review of Machine Learning Techniques for Analog Integrated Circuit Design and Optimization
by Vasile Grosu
Journal of Engineering Research and Sciences, Volume 5, Issue 9, Page # 15-34, 2026; DOI: 10.55708/js0509002
Abstract: Analog electronic design automation has historically lagged behind its digital counterpart due to the complex, multi-dimensional design space and strict sensitivity to parasitic effects, process variations, and geometric constraints. A comprehensive review of the evolution, current state-of-the-art, and future trajectories of electronic design automation tools specifically tailored for analog integrated circuit design could serve as… Read More
(This article belongs to the Section Electronic Engineering (EEE))
Fleet-Based Reliability Framework for Low-Voltage Electrical Subsystems in Autonomous Vehicles
by Shrikant Pawar
Journal of Engineering Research and Sciences, Volume 5, Issue 9, Page # 35-60, 2026; DOI: 10.55708/js0509003
Abstract: The low-voltage (LV) electrical subsystem of an autonomous vehicle (AV), spanning the auxiliary battery and management system, DC-DC conversion, power distribution, compute ECUs, in-vehicle communication networks, and harnesses, must operate continuously; its failures are a leading cause of in-service downtime even when propulsion is healthy. Yet no method connects fleet-derived reliability evidence to the assurance… Read More
(This article belongs to the Section Robotics (ROB))