Benoit Favre

CPT+SFT may lead to the best performance in multiple-choice QA, but SFT alone is often a surprisingly effective and economical choice.

Finetuning speech recognition models can either amplify or erase critical speaker group information, depending on the training focus and fairness interventions.

LLMs judging medical QA are heavily biased by the answer's source, but a little fine-tuning can make smaller models surprisingly fair.