ROMay 9

ATAAT: Adaptive Threat-Aware Adversarial Tuning Framework against Backdoor Attacks on Vision-Language-Action Models

arXiv:2605.0861261.0
Predicted impact top 34% in RO · last 90 daysOriginality Incremental advance
AI Analysis

For security researchers and practitioners, this work reveals a critical vulnerability in VLA models and provides a method for effective backdoor attacks, though it is incremental as it adapts existing adversarial tuning concepts.

This paper addresses backdoor attacks on Vision-Language-Action (VLA) models, identifying 'Gradient Interference' as a key obstacle. The proposed ATAAT framework achieves over 80% Targeted Attack Success Rate with only 5% poisoning rate while maintaining stealthiness.

Addressing the escalating security vulnerabilities in Vision-Language-Action (VLA) models, this study investigates backdoor attacks targeting the visual pathway. We identify a core obstacle causing the failure of traditional attack paradigms: "Gradient Interference." This phenomenon represents an optimization failure triggered by conflicting strategies during end-to-end training. To resolve this, we propose an Adaptive Threat-Aware Adversarial Tuning (ATAAT) framework. Through its core "Threat-Method Adaptive Mapping" mechanism, ATAAT intelligently selects the optimal gradient decoupling strategy based on the adversary's capabilities. Extensive experiments demonstrate that ATAAT exhibits significant advantages, achieving a highly robust Targeted Attack Success Rate (TASR > 80%) while maintaining extreme stealthiness with merely a 5% poisoning rate. It efficiently handles complex semantic-level triggers and achieves implicit decoupled attacks in data poisoning scenarios for the first time. This work reveals a critical security vulnerability in VLAs and provides theoretical and methodological support for future defense architectures.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes