Publication Date
Fall 2025
Degree Type
Master's Project
Degree Name
Master of Science in Computer Science (MSCS)
Department
Computer Science
First Advisor
William Andreopoulos
Second Advisor
Thomas Austin
Third Advisor
Navrati Saxena
Keywords
Llama3, Llama2, Gemma, Resume
Abstract
Applicants often use generalized bullet points in resumes, which do not emphasize the skills and accomplishments relevant to a particular job listing, and thus reduce their chances of being selected by recruiters and Applicant Tracking Systems. This paper presents a way of automation that can be used to complement the existing resume point bullets by adding job-specific keywords, quantifiable measurements, and action verbs specific to the software-engineering job. We train and compare three large language models: Llama 3 8B, Llama 2 7B, and Gemma 7B using effective training regimes. The quality and relevance are estimated through a complex set of measures (ROUGE, BLEU, METEOR). Therefore, this work will provide job applicants with effective tools to build stronger, personalized resumes that are more in line with job requirements.
Recommended Citation
Vedant, Gupta, "Resume Bullet Point Enhancement Using Multi-Model Fine-Tuning: A Comparative Study" (2025). Master's Projects. 1609.
DOI: https://doi.org/10.31979/etd.ktqg-7z58
https://scholarworks.sjsu.edu/etd_projects/1609
