Publication Date
Spring 2026
Degree Type
Master's Project
Degree Name
Master of Science in Computer Science (MSCS)
Department
Computer Science
First Advisor
Navrati Saxena
Second Advisor
Amith Kamath Belman
Third Advisor
Kalindi Parekh
Keywords
Retrieval-Augmented Generation (RAG), Large Language Models (LLM), Evaluator-Optimizer Framework, Agentic RAG, RAGAS
Abstract
Language based video reasoning (LBVR) systems rely on Retrieval-Augmented Generation (RAG) over transcripts and captions of videos. However, naive single-pass RAG pipelines are highly sensitive to embedding and retrieval choices, hence they are prone to hallucination especially when dealing with technical and reasoning corpus. This constrains developers to utilize computationally expensive embedding models and retrieval strategies. This work investigates an Evaluator-Optimizer RAG framework specifically designed for academic reasoning and research focused LBVR. The framework audits and improves a first pass answer by validating logical consistency, grounding, and relevance. We implement a cross-reflection approach to avoid confirmation bias by using GPT-4o for first pass generation and Gemini-3-Pro for feedback refinement. We compare this framework with baseline implementations integrated with state-of-the-art (SOTA) embedding model (BGE). This study demonstrates that feedback-based cross-reflection with lightweight embedding model (MiniLM) performs on par with baselines, enabling implementation of research grade LBVR systems in low compute environments. For evaluation we utilized RAGAS framework rather than static n-grams metrics to retain the semantic meaning of the responses and discussing the trade-offs between faithfulness and answer accuracy. Our results indicate that cross-reflective feedback driven LBVR RAG systems for academia can reduce the dependence on compute resources opening opportunities for deployment in resource-constraint settings such as edge devices.
Recommended Citation
Atreya, Udayan, "Cross-Reflection Evaluator-Optimizer Feedback Retrieval-Augmented Generation (RAG) Framework" (2026). Master's Projects. 1787.
DOI: https://doi.org/10.31979/etd.3psf-m7aj
https://scholarworks.sjsu.edu/etd_projects/1787
