InfoResearchPreprint
IDEA: Intent-Dominance Evidence Audit for Inference-Time Backdoor Detection in GUI Agents
- Published
- Record updated
Summary
The paper proposes IDEA (Intent-Dominance Evidence Audit), a method for detecting backdoor activation in open-weight GUI agents at inference time, per step, without trigger knowledge or attack data. It combines Intent Counterfactual Probing (ICP) and Region Causal Audit (RCA) in a gated conformal score with a finite-sample false-positive guarantee. Experiments on Qwen2.5-VL-3B across five attack families show it detects backdoors under a controlled false-positive rate, outperforms prior detectors, and localizes the trigger region for each alarm.
Related items
- InfoThreatsDay: GPT-6.2 Jailbreak Flaws, BubbleLoader PNG Malware, $500M Ransom Demands + 17 More StoriesSimilar attack · The Hacker News
- MediumCVE-2026-128136: NLTK pickle loading lets crafted models read or truncate local filesSimilar attack · NVD/CVE Database
- Info30 Years of Gatekeeper Asymmetry Doom: The Paradox that Makes Powerful LLMs More VulnerableSimilar attack · Check Point Research
- HighCVE-2026-76818: Axon rtkx bootstrap runs unverified downloaded binarySimilar attack · NVD/CVE Database
- InfoJailbreak Detection on Large Language Models via Contrastive Prototype StructuresSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)