8 papers · 1 filter
Blocking at the Boundary: Auditing Long-Horizon Agents against Staged Prompt Injection
Jingkai Liu, Yufei Han, Xiaoting Lyu +2
Long-horizon agents consume external content, invoke tools, and modify persistent state. Indirect prompt injection can exploit task-specific context, propagate across causally conn…
Conformal Privacy Auditing: Calibrated Re-identification Attacks with Statistical Guarantees
Shuo Huang, Gholamreza Haffari, Xingliang Yuan +2
Empirical identity leakage from released text is increasingly driven by attackers that combine large language models (LLMs) with auxiliary knowledge to link documents to individual…
Distributed and Private Textual Data Synthesis from Embeddings
Ergute Bao, Hongyan Chang, Ali Shahin Shamsabadi +2
We revisit differentially private (DP) text synthesis in the realistic setting of distributed users, where privacy concerns preclude a trusted curator with access to raw user texts…
Low-ASR Backdoors: Exploiting Attack Success Rate Reduction and Attacker-Defender Asymmetry
Arham Riaz, Ting Yu
Backdoor attacks are among the most effective and stealthy attacks in deep learning. Existing attacks and defenses are largely designed and evaluated under the assumption that succ…
Overcoming the Retrieval Barrier: Indirect Prompt Injection in the Wild for LLM Systems
Hongyan Chang, Ergute Bao, Xinjian Luo +1
Large language models (LLMs) increasingly rely on retrieving information from external corpora. This creates a new attack surface: indirect prompt injection (IPI), where hidden ins…
Exposing Hidden Interfaces: LLM-Guided Type Inference for Reverse Engineering macOS Private Frameworks
Arina Kharlamova, Youcheng Sun, Ting Yu
Private macOS frameworks underpin critical services and daemons but remain undocumented and distributed only as stripped binaries, complicating security analysis. We present MOTIF,…
From Text to Actionable Intelligence: Automating STIX Entity and Relationship Extraction
Ahmed Lekssays, Husrev Taha Sencar, Ting Yu
Sharing methods of attack and their effectiveness is a cornerstone of building robust defensive systems. Threat analysis reports, produced by various individuals and organizations,…
DeBackdoor: A Deductive Framework for Detecting Backdoor Attacks on Deep Models with Limited Data
Dorde Popovic, Amin Sadeghi, Ting Yu +2
Backdoor attacks are among the most effective, practical, and stealthy attacks in deep learning. In this paper, we consider a practical scenario where a developer obtains a deep mo…