BEAT: Visual Backdoor Attacks on VLM-based Embodied Agents via Contrastive Trigger Learning

Documenti analoghi

• MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Poisoning Attacks
  di: Ha, Hyeonjeong, et al.
  Pubblicazione: (2025)
• InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents
  di: Zhan, Qiusi, et al.
  Pubblicazione: (2024)
• IAG: Input-aware Backdoor Attack on VLM-based Visual Grounding
  di: Li, Junxian, et al.
  Pubblicazione: (2025)
• EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents
  di: Yang, Rui, et al.
  Pubblicazione: (2025)
• BackdoorVLM: A Benchmark for Backdoor Attacks on Vision-Language Models
  di: Li, Juncheng, et al.
  Pubblicazione: (2025)