A test-time meta-learning approach that steers model behavior via in-context vector arithmetic, offering a practical alternative to few-shot ICL under strict context budgets. Includes a reproducible codebase and open-source toolkit (in progress).
An analysis suite for quantifying LLM sensitivity to demonstration ordering and placement in prompts. Supports large-scale evaluations across multiple model families and tasks, with distilled prompt-design recommendations.
A reproducible codebase and synthetic dataset combining real and LLM-generated biased text in African American English (AAE), enabling automated fairness evaluation, ablation studies, and dialect-aware benchmarking.
A cross-platform mobile app enabling churches to stream services live, schedule events, and engage communities in a single unified workflow. Features push notifications, content management, and social sharing.
A sensor-fusion pipeline combining smartphone accelerometer/gyroscope signals with an ankle-band sensor to capture lower-body dynamics missed by phone-only trackers. Includes feature extraction and classification models with controlled ablation experiments.