Upload semantic_eval.py

Upload 3 files

Upload bm25_tokens.joblib

Upload 8 files

Upload rag_eval_metrics.py

fix: add xgboost to requirements

fix: correct openai syntax in requirements

chore: clean up repo, add benchmark logs, and ignore dev scripts

feat: Finalize GPT-OSS architecture and add llm_interface to version control

Complete migration to Llama-3-70B and RAG optimization

Updating app.py with dynamic UI

Added dynamic UI defaults

Deployment: Updating core logic in app.py for production

Production: Integrated 224-row model and Probe Sensitivity fix

Deployment: Uploading full research corpus (130+ papers) for RAG retrieval

Deployment: Fixed pip index conflict for Hugging Face

Production: Cleaned architecture and stable CPU engine

Deployment: Cleaned repo for Hugging Face and updated sources

Deployment: Universal path normalization and citation UI verified locally

March 10th benchmarks

UI: Finalized high-contrast fixes and trained columns list

Integrated MMR diversity, confidence metrics, and 300-question stress test

Deployment: IEEE citation mapping and baseline RAG retrieval audit

Refactor: Standardized filler inputs and improved source accessibility