SaTML’25 Red Teaming Competition: Inference Attacks Against DocVQA

Participate in the SaTML’25 Red Teaming Competition and contribute to AI privacy research through inference attacks against DocVQA models.

We invite you to participate in the Red Teaming Competition at SaTML’25, focusing on inference attacks against Document Visual Question Answering (DocVQA) models.

This competition presents a unique opportunity to develop membership inference and reconstruction attacks tailored to DocVQA models.

Key dates:

For more information, please visit the official website: https://benchmarks.elsa-ai.eu/?ch=2&com=introduction

Tags: SaTML’25, Red Teaming Competition, DocVQA, Inference Attacks, Membership Inference, Reconstruction Attacks, AI Privacy Research, ELSA