4 papers
The Heterogeneous Safety Impacts of Benign Multilingual Fine-Tuning
Will Hawkins, Kaivalya Rawal, Jonathan Rystrøm +8
Fine-tuning a large language model is a ubiquitous method for enhancing its capability on a specific downstream task. However, prior work has shown that this increase in capability…
Toward Calibrated, Fair, and accurate Deepfake Detection
Ryan Brown, Chris Russell
Deepfake detectors show large performance gaps across demographic groups. Existing fairness approaches require demographic labels, retraining, or sacrifice accuracy. We introduce F…
Adaptive Calibration for Fair and Performant Facial Recognition
Ryan Brown, Chris Russell
We introduce Adaptive Calibration (AC), a novel calibration strategy for facial recognition that maps cosine similarity between normalized embeddings to well-calibrated probabiliti…
Task-Specific Knowledge Distillation via Intermediate Probes
Ryan Brown, Chris Russell
Knowledge distillation from large language models (LLMs) assumes that the teacher's output distribution is a high-quality training signal. On reasoning tasks, this assumption is fr…