Advanced ML & Bias Detection Research

[adhit-r]

Description

Research and implement advanced bias detection and LLM integration capabilities.

Scope

• Advanced bias detection methods (causal inference, intersectional fairness metrics)
• LLM-specific bias evaluation (prompt-level bias, output distribution analysis)
• Multi-model comparison and benchmarking for fairness
• Research integration — keep up with latest fairness literature and tooling

Consolidates

#59, #60

[IlyasFardaouix]

I've taken a closer look at the discussion around advanced bias detection and LLM integration, and one thing that caught my attention is the lack of concrete implementation details in the main.py file. Specifically, I've noticed that the lifespan function in main.py doesn't seem to be handling any bias detection or LLM integration logic. Could you provide more context on how you envision this integration playing out in the lifespan function or elsewhere in the codebase?