statistical-evaluation

frontier-evals-harness is a lightweight framework for benchmarking frontier language models. It provides deterministic suite versioning, modular adapters, standardized scoring, and paired statistical comparisons with confidence intervals. Built for regression tracking and analysis, it enables reproducible evaluation without infrastructure.

Quantitative performance benchmarking of machine learning classifiers for phishing detection, utilizing precision, recall, and F1-score optimization on high-dimensional labeled datasets.

DTU 02445 individual assignment: subject-level evaluation of regression models on heart-rate data

Classification models for detecting fake reviews and predicting software bugs. Includes implementations of decision trees, bagging, random forests, logistic regression, and Naive Bayes, with statistical evaluation using McNemar's test.

Statistical evaluation of Gemma refusal robustness using direction ablation and official SORRY-Bench scoring.

Provider-neutral AI evaluation toolkit for reusable test cases, regression datasets, provider adapters, deterministic and LLM judges, agent trajectory and security evaluation, statistical analysis, quality/cost/latency trade-offs, online experiments, evidence guardrails, and release gates.