arXiv subjects

Pranav S

Publications and source records attributed to Pranav S.

4 recordsLinked to original sources

arXiv2026

An Explainable Two Stage Deep Learning Framework for Pericoronitis Assessment in Panoramic Radiographs Using YOLOv8 and ResNet-50

Objectives: To overcome challenges in diagnosing pericoronitis on panoramic radiographs, an AI-assisted assessment system integrating anatomical localization, pathological classification, and interpretability. Methods: A two-stage deep learning pipeline was implemented. The first stage used YOLOv8 to detect third molars and classify their anatomical positions and angulations based on Winter's classification. Detected regions were then fed into a second-stage classifier, a modified ResNet-50 architecture, for detecting radiographic features suggestive of pericoronitis. To enhance clinical trust, Grad-CAM was used to highlight key diagnostic regions on the radiographs. Results: The YOLOv8 component achieved 92% precision and 92.5% mean average precision. The ResNet-50 classifier yielded F1-scores of 88% for normal cases and 86% for pericoronitis. Radiologists reported 84% alignment between Grad-CAM and their diagnostic impressions, supporting the radiographic relevance of the interpretability output. Conclusion: The system shows strong potential for AI-assisted panoramic assessment, with explainable AI features that support clinical confidence.

cs.CV↗

arXiv2025

Mathematical Framework for Custom Reward Functions in Job Application Evaluation using Reinforcement Learning

Most of the traditional Applicant Tracking Systems (ATS) depend on strict matching using keywords, where candidates that are highly qualified are many times disqualified because of minor semantic differences. In this article, the two-stage process of developing a more comprehensive resume assessment system based on a small language model that is trained with fewer than 600M parameters is introduced and fine-tuned by using GRPO with a uniquely designed reward function. The initial stage is Supervised Fine-Tuning (SFT), which is used to create a strong base model with the ability to perceive resumes beyond superficial overlap of keywords. This SFT model is further optimized in the second step with Reinforcement Learning (RL) via GRPO with the help of multi-component-based rewarding, which will not be considered as a commission of tokens matching. In the initial RL experiments, we found a severe difficulty in the shape of reward hacking: overly aggressive penalty terms resulted in unstable training dynamics and prohibitively negative model behavior. This was solved by trial-and-error refinement of the reward and careful training hyperparameter tuning, which led to a stable and controlled process of gentle polishing. The GRPO-refined model shows high real-life performance, as it shows an accuracy of 91% on unseen data used for testing. It has a high recall of 0.85 on the SELECTED class with a perfect precision of 1.0, which highlights its high reliability for identifying qualified applicants. These findings demonstrate that an appropriately structured two-step fine-tuning pipeline can effectively be used to transfer a small language model into human-like candidate evaluation, surpassing the shortcomings of both traditional ATS systems and unrefined uses of reinforcement learning.

cs.LG↗

arXiv2024

Modified RRT* for Path Planning in Autonomous Driving

Essential tasks in autonomous driving includes environment perception, detection and tracking, path planning and action control. This paper focus on path planning, which is one of the challenging task as it needs to find optimal path in highly complex and dynamic environments. Usually, a driving scenario has large number of obstacles in their route. In this paper, we propose a two-stage path planning algorithm named Angle-based Directed Rapidly exploring Random Trees (AD-RRT*) to address the problem of optimal path in complex environment. The proposed algorithm uses A* algorithm for global path planning and modifies RRT* to bound the samples using angle. The efficiency of the proposed algorithm is evaluated through experiments in different scenarios based on the location and number of obstacles. The proposed algorithm showed higher rate of convergence with reduced time and less number of nodes than the base RRT* algorithm.

cs.RO↗

arXiv2023

Enhancing Room Security and Automating Class Attendance Using ID Cards

With the rapid advancements in technology, automation has emerged as the future of human endeavors. From simple tasks like attendance management to complex security systems, automation has the potential to revolutionize various aspects of our lives. This research paper explores the implementation of a method aimed at enhancing room security in hostels and automating class attendance using ID cards. In this study, we propose a system that utilizes the unique identity information stored in ID cards for various security and check-in tasks. By integrating RFID (Radio-Frequency Identification) reader technology, GSM modules, Node MCU, and Arduino, we create a comprehensive solution. The RFID reader scans the ID card, extracting the relevant information and verifying the user's identity. The data is then transmitted via the GSM module to a central database, ensuring real-time monitoring and security measures. Moreover, the system also enables the automation of class attendance. By utilizing the same ID cards, students can simply tap their cards on a reader placed in the classroom. This information is recorded automatically, eliminating the need for manual attendance taking and reducing errors and time consumption. This research project highlights the practical implementation of ID card technology to enhance room security in hostels and automate class attendance processes. By leveraging the power of automation, we aim to streamline administrative tasks, improve security measures, and optimize efficiency in educational institutions and other relevant settings.

cs.CR↗