Paper titled “Rethinking Dataset Distillation: Hard Truths about Soft Labels: has been selected as a Best Paper Finalist at CVPR 2026!
16 Jun 26 Web Admin
We are excited to share that our paper 𝗥𝗲𝘁𝗵𝗶𝗻𝗸𝗶𝗻𝗴 𝗗𝗮𝘁𝗮𝘀𝗲𝘁 𝗗𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻: 𝗛𝗮𝗿𝗱 𝗧𝗿𝘂𝘁𝗵𝘀 𝗮𝗯𝗼𝘂𝘁 𝗦𝗼𝗳𝘁 𝗟𝗮𝗯𝗲𝗹𝘀 has been selected as a 𝗕𝗲𝘀𝘁 𝗣𝗮𝗽𝗲𝗿 𝗙𝗶𝗻𝗮𝗹𝗶𝘀𝘁 𝗮𝘁 𝗖𝗩𝗣𝗥 𝟮𝟬𝟮𝟲!
Out of more than 𝟭𝟲,𝟬𝟬𝟬 𝘀𝘂𝗯𝗺𝗶𝘀𝘀𝗶𝗼𝗻𝘀, the paper was recognized among the 𝗧𝗼𝗽 𝟭𝟱 𝗽𝗮𝗽𝗲𝗿𝘀, an incredible honor and a testament to the hard work and dedication of the team.
Dataset distillation has emerged as a promising direction for efficient learning, and this work revisits some of its core assumptions, uncovering important insights about the role of soft labels and their limitations.
Congratulations to all the authors on this outstanding achievement!
