arXiv AI July 29, 2026 By Maria Rosaria Briglia, Igor Maljkovic, Antonio Emanuele Cin\`a, Luca Oneto, Iacopo Masi, Fabio Roli

Architectural Backdoors in Vision-Language Model Supply Chains via Representation Steering

Read the original on arXiv AI →

arXiv:2607. 25479v1 Announce Type: cross Abstract: Vision--Language Models (VLMs) are increasingly deployed through a model supply chain in which pretrained checkpoints, architecture definitions, text encoders, and exported computation graphs are distributed by third parties and reused across downstream services.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv AI

Jul 1

A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems

arXiv:2606. 31639v1 Announce Type: cross Abstract: Large language models are no longer only text generators.

By Seyed Bagher Hashemi Natanzi, Bo Tang

llms agents robotics safety

arXiv Machine Learning

Jun 3

BYORn: Bootstrap Your Own Responses to Defend Large Vision-Language Models Against Backdoor Attacks

arXiv:2606. 02947v1 Announce Type: new Abstract: Supervised fine-tuning is the predominant approach for adapting autoregressive vision-language models to downstream tasks.

By Ivan Saboli\'c, Marin Or\v{s}i\'c, Josip \v{S}ari\'c, Sven Lon\v{c}ari\'c

llms fine-tuning multimodal

arXiv Machine Learning

Jul 21

BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors

arXiv:2501. 02373v3 Announce Type: replace Abstract: Task arithmetic in large-scale pre-trained models enables agile adaptation to diverse downstream tasks without extensive retraining.

By Chia-Yi Hsu, Yu-Lin Tsai, Yu Zhe, Yan-Lun Chen, Chih-Hsun Lin, Chia-Mu Yu, Yang Zhang, Chun-Ying Huang, Jun Sakuma

arXiv Machine Learning

Jul 7

VLMGuard: Bootstrapping Malicious Prompt Detectors from Unlabeled Vision-Language Prompts in the Wild

arXiv:2410. 00296v2 Announce Type: replace Abstract: Vision-language Models (VLMs) are essential for contextual understanding of both visual and textual information.

By Junlin Fang, Wenyu Chen, Reshmi Ghosh, Robert Sim, Ahmed Salem, Vitor R. Carvalho, Emily Lawton, Sharon Li, Jack W. Stokes, Sean Du

llms multimodal benchmarks

arXiv AI

Jul 7

VISOR: Visual Input-based Steering for Output Redirection in Vision-Language Models

arXiv:2508. 08521v2 Announce Type: replace-cross Abstract: Vision Language Models (VLMs) are increasingly being used in a broad range of applications, bringing their security and behavioral control to the forefront.

By Mansi Phute, Ravikumar Balakrishnan

llms robotics multimodal benchmarks safety

arXiv Machine Learning

Jun 30

High-Entropy Tokens as Multimodal Failure Points in Vision-Language Models

arXiv:2512. 21815v4 Announce Type: replace-cross Abstract: Vision-language models (VLMs) achieve remarkable performance but remain vulnerable to adversarial attacks.

By Mengqi He, Xinyu Tian, Xin Shen, Jinhong Ni, Shu Zou, Zhaoyuan Yang, Jing Zhang

llms multimodal safety

Related stories

A Lifecycle and Application-Stack Survey of Large Language Model Vulnerabilities: Attacks, Risks, Defenses, and Open Problems

BYORn: Bootstrap Your Own Responses to Defend Large Vision-Language Models Against Backdoor Attacks

BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors

VLMGuard: Bootstrapping Malicious Prompt Detectors from Unlabeled Vision-Language Prompts in the Wild

VISOR: Visual Input-based Steering for Output Redirection in Vision-Language Models

High-Entropy Tokens as Multimodal Failure Points in Vision-Language Models