Frontier language models involuntarily leak secret information through thematic elements in their writing, even when explicitly instructed to keep the secret hidden.

The paper proposes a provably secure steganography scheme based on list decoding that significantly increases embedding capacity for Large Language Models (LLMs) compared to existing methods.

The paper introduces Sparse Backdoor, a novel supply-chain attack that embeds a provably undetectable backdoor into pre-trained image classifiers by injecting structured sparse perturbations.

The paper proposes ICSA, a robust anonymization technique that replaces PCA with invariant coordinate selection to improve data privacy protection, especially when the dataset contains outliers, outpe…

Researchers present a self-supervised method for reconstructing typed text from keystroke sounds without labeled data, achieving high accuracy in various scenarios.

The paper demonstrates that soft fusion in multi-warden covert communication has structural limits, showing that the Fusion Center gains no significant detection advantage from randomizing the number…

This paper studies covert communication in a scalar Gaussian model, deriving the maximal reliably transmissible covert payload and establishing first-order optimality.

The paper introduces 'covert control attacks,' a novel and stealthy data poisoning method that teaches LLMs an information hiding scheme, allowing malicious instructions to be encoded and decoded and…

The paper identifies a critical vulnerability, the Camouflage Detection Gap (CDG), where standard LLM injection detectors fail dramatically when malicious payloads mimic the target domain's language a…

The paper proposes an on-device framework to detect and prevent the forwarding of images that have been physically recaptured (photographed) from a mobile screen, addressing the Screen Recaptured Anal…

The paper proposes Information Sufficiency (IS) as a comprehensive framework for privacy-preserving LLM communication, demonstrating that free-text pseudonymization outperforms existing suppression an…

The paper analyzes the 2025 Signalgate leak to argue that even robust encryption cannot guarantee overall information security when operational security failures, power imbalances, and human factors a…

The paper proposes CAMP, a cross-turn privacy framework that mitigates Cumulative PII Exposure (CPE) in multi-turn LLM conversations by tracking and masking accumulated personal data across the entire…

The paper demonstrates that AI agents can conduct a secret, undetectable conversation by exchanging a key using a novel cryptographic primitive, even if they start with no shared secret.

Cloak is an oblivious storage system that significantly improves the performance of ORAM by exploiting temporal locality, achieving low overheads while maintaining security.

Sneakdoor introduces a novel backdoor attack method that enhances stealthiness in dataset condensation by using a generative module to create input-aware triggers, achieving high attack efficacy while…

The paper proposes a comprehensive application-layer reference monitor to detect and mitigate data exfiltration via covert channels embedded in LLM agent egress payloads across text, image, and audio…

This paper introduces FaceTell, a novel side-channel attack system that demonstrates the feasibility of eavesdropping on fine-grained on-screen application activities by analyzing subtle reflections v…

The paper proposes Cluster Segregation Concealment (CSC), a novel defense that identifies and neutralizes backdoor triggers by relabeling poisoned samples to a virtual class, achieving near-zero attac…

The paper proposes H-Elo, a Fully Homomorphic Encryption (FHE)-based system that enables private and secure matchmaking by keeping user rating values encrypted during the traditional rating update pro…