order-sensitivity

Official repository for Facet-Probe, an evaluation framework for auditing order sensitivity in multimodal large language models. You can find the corresponding pre-print here: https://arxiv.org/abs/2606.26079

ChoiceBench is a lightweight framework for MCQ evaluation-method research on LLMs, with built-in support for answer-order bias analysis and mitigation methods.

Stateful agent-interface benchmark for rewrite vs stable-line patch reliability in ordering agents.