VEglue: Testing Visual Entailment Systems via Object-Aligned Joint Erasing

no code implementations • 5 Mar 2024 • Zhiyuan Chang, Mingyang Li, Junjie Wang, Cheng Li, Qing Wang

Visual entailment (VE) is a multimodal reasoning task consisting of image-sentence pairs whereby a promise is defined by an image, and a hypothesis is described by a sentence.

Multimodal Reasoning Sentence +1

Adversarial Testing for Visual Grounding via Image-Aware Property Reduction

no code implementations • 2 Mar 2024 • Zhiyuan Chang, Mingyang Li, Junjie Wang, Cheng Li, Boyu Wu, Fanjiang Xu, Qing Wang

To this end, we propose PEELING, a text perturbation approach via image-aware property reduction for adversarial testing of the VG model.

Visual Grounding

Play Guessing Game with LLM: Indirect Jailbreak Attack with Implicit Clues

no code implementations • 14 Feb 2024 • Zhiyuan Chang, Mingyang Li, Yi Liu, Junjie Wang, Qing Wang, Yang Liu

With the development of LLMs, the security threats of LLMs are getting more and more attention.