SPARTA: Evaluating Reasoning Segmentation Robustness through Black-Box Adversarial Paraphrasing in Text Autoencoder Latent Space

Viktoriia Zinkovich, Anton Antonov, Andrei Spiridonov +6

Multimodal large language models (MLLMs) have shown impressive capabilities in vision-language tasks such as reasoning segmentation, where models generate segmentation masks based…