Synthetic Dataset and Evaluation Results: Validating AI-Generated Content: Security Challenges and Solutions Inspired by OWASP

Synthetic Dataset and Evaluation Results for LLM Output Validation This repository contains the synthetic dataset and evaluation results used in the Master's thesis: Validating AI-Generated Content: Security Challenges and Solutions Inspired by OWASP by Kamran Khan, Tampere University, 2025. Contents 1. Dataset dataset.csv: Synthetic dataset generated for evaluating LLM outputs Contains 446 samples across 9 categories Each sample includes: Context, LLMOutput, TrueLabel Generated using: Deepseek-V3.2-Exp , Gpt-4 2. Evaluation Results BaselineResulat.csv: Results from the baseline model using only procedural security and schema/structure validator agent(non-llm) NoJudgeAgentResult: Results ablation study using semantic, validator and Semantic Agent without Judge audit EvalauationResult: Results from comple