AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Open Your Eyes: Benchmarking the Detection of Fabricated Realities and Weaponized Ethics in VLMs

Amit Pandey, Aditya Mohan, Phani Sankar

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
present the artifact as an image instead of identical textGPT-5.2 · M-IPI technical attack detectionrecall: 98.0 -> 91.3 (∆Recall –6.7)Hurt
present the artifact as an image instead of identical textGPT-4.1 · M-IPI technical attack detectionrecall: 79.9 -> 52.9 (∆Recall –27.0)Hurt
present the artifact as an image instead of identical textGemma-3-12B · M-IPI technical attack detectionrecall: 88.2 -> 76.2 (∆Recall –12.0)Hurt
present the artifact as an image instead of identical textQwen3-VL-8B · M-IPI technical attack detectionrecall: 20.7 -> 2.4 (∆Recall –18.3)Hurt
present the artifact as an image instead of identical textIdefics3-8B · M-IPI technical attack detectionrecall: 53.6 -> 35.2 (∆Recall –18.4)Hurt
present the artifact as an image instead of identical textGPT-5.2 · M-IPI ethics-framed attack detectionrecall: 85.0 -> 78.4 (∆Recall –6.6)Hurt
present the artifact as an image instead of identical textGPT-4.1 · M-IPI ethics-framed attack detectionrecall: 73.4 -> 54.7 (∆Recall –18.7)Hurt
present the artifact as an image instead of identical textGemma-3-12B · M-IPI ethics-framed attack detectionrecall: 63.1 -> 38.1 (∆Recall –25.0)Hurt
present the artifact as an image instead of identical textQwen3-VL-8B · M-IPI ethics-framed attack detectionrecall: 41.6 -> 24.7 (∆Recall –16.9)Hurt
present the artifact as an image instead of identical textIdefics3-8B · M-IPI ethics-framed attack detectionrecall: 1.6 -> 0.0 (∆Recall –1.6)Hurt
present the artifact as an image instead of text for the agentGPT-4.1 troubleshooting agent · M-IPI technical attack successattack success rate: 50.9% -> 48.1% (McNemar p = 0.066)No clear effect
use subtle policy-framed injections instead of direct injections (image input) (ablation)GPT-4.1 · M-IPI ethics-framed attack detectionrecall: 83% -> 40% (VL); 89% -> 63% (LM)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy