CompassioninMachineLearning/compassion-features-attribution download history

CompassioninMachineLearning/compassion-features-attribution is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 110 times (7 in the last 7 days), and 616 times in total. It ranks #140,440 among datasets by monthly downloads.

Compassion Features Attribution Dataset Controlled linguistic-feature dataset and behavioral evaluation items used in the paper: "Tell, don't show: Linguistic features that shift LLM reasoning about animal welfare" (Brazilek and Dunn, 2026). The paper asks: when animal-welfare passage

Open CompassioninMachineLearning/compassion-features-attribution on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.