Tag: Safety

All the articles with the tag "Safety".

• Adversarial Attacks in Multimodal Systems: A Practitioner's Survey

  Published: 8 May, 2025 |  at  11:09 AM

  

  96.71 🤔

  This survey paper provides a comprehensive overview of adversarial attacks on multimodal AI systems across text, image, video, and audio modalities, categorizing threats by attacker knowledge, intention, and execution to equip practitioners with knowledge of vulnerabilities and cross-modal risks.

• Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs

  Published: 8 May, 2025 |  at  11:07 AM

  

  94.57 🤔

  This paper proposes a three-dimensional taxonomy and develops TTP and HarmFormer tools to filter harmful content from web-scale LLM pretraining datasets, revealing significant toxicity prevalence and persistent safety gaps through benchmarks like HAVOC.

• Enhancing Safety Standards in Automated Systems Using Dynamic Bayesian Networks

  Published: 8 May, 2025 |  at  11:06 AM

  

  90.63 🤔

  This paper proposes a Dynamic Bayesian Network framework for autonomous vehicles that enhances safety in cut-in maneuvers by integrating lateral evidence and probabilistic safety assessments, achieving superior crash avoidance in high-speed scenarios (9.22% crash rate) compared to baseline models in the JRC-FSM simulator.

• MELON: Provable Indirect Prompt Injection Defense via Masked Re-execution and Tool Comparison

  Published: 8 May, 2025 |  at  10:22 AM

  

  89.40 🤔

  MELON introduces a novel training-free defense against indirect prompt injection attacks on LLM agents by detecting independence of tool calls from user inputs through masked re-execution, achieving superior attack prevention (0.24% ASR on GPT-4o) and utility preservation (58.78% UA on GPT-4o) compared to existing methods.

• Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks

  Published: 8 May, 2025 |  at  11:07 AM

  

  87.75 🤔

  ASTRA introduces an efficient defense for Vision Language Models by adaptively steering activations away from adversarial directions using image attribution, achieving state-of-the-art performance in mitigating jailbreak attacks with minimal impact on benign utility and high inference efficiency.