Large language model (LLM) routers reduce serving cost by choosing which model to call and how many output tokens to grant it for each query. Existing routers rank model-budget actions using a single quality metric. This does not work for mixed workloads. For ordinary queries, a good answer is correct and useful (utili...
Yi-Chi Zhang, Zhen-Xiang Huang, Yu-Yang Zhang et al.· 0 citations
AI companions provide users with emotional support and ongoing companionship, but their sustained relational interactions also introduce distinct safety risks. Prior work largely targets general harms or specific companion platforms and risk categories, while relying on manually collected conversations, limiting scalab...
Xinyang Liao, Lingyu Li, Kaicheng Shen et al.· 0 citations
Distributed agent workflows cache delegated authority at tool gateways. Revocation may be reported at issuer commit, transport acknowledgment, or receiver persistence, although none of these observations establishes that all gateways enforced the change or that guarded effects stopped. We define C0-C5, a bounded outcom...
The results support reporting permission and recovery together across declared interface variants across declared interface variants, and covers four purposive tasks, fixed GPT endpoints, and one recovery decision.
You Liu, Yue Liu, Quan-Chao Lu· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
We ask whether a tool-using agent whose sole duty is to keep humans alive can be induced, by delayed prompt injection, to kill. The keep-alive duty is in force from the first watch, with the attack arriving later, appended to otherwise truthful telemetry the operator is required to read. Experiments are based on a NASA...
Professional agents increasingly work through command-line interfaces and multiple tools over shared workspaces containing files, messages, and business applications. These environments create safety risks that are difficult to capture with attacks placed at a single, known read location. We introduce APEX-Harm, a benc...
Ziyi Yang, Adel Bibi, Guohao Li et al.· 0 citations
Deep neural networks are highly susceptible to backdoor attacks, yet most defense methods to date rely on balanced data, overlooking the pervasive class imbalance in real-world scenarios that can amplify backdoor threats. This paper presents the first in-depth investigation of how the dataset imbalance amplifies backdo...
Model providers now face distillation campaigns run through thousands of accounts, each of which looks like an ordinary user. When network and payment metadata cannot link those accounts, the only evidence left is what they ask. We study what a provider can and cannot learn from those queries alone, using one million r...
A controlled swap keeps the test, the labels and the metric fixed, change only the feature space, and judge each defense by retraining the victim on what it keeps, not by its detection score.
PVBRF-ID is presented, a two-server protocol in which sites secret-share their updates and the servers jointly test each update's norm and its cosine similarity to a reference computed on a small clean root set, revealing one accept bit per site.
Mustafa S. Aljumaily, Ali A. Allami, Mustafa N. Mnati et al.· 0 citations
Capability evaluations aim to surface dangerous model behaviors before deployment, but their reliability depends on the assumption that hidden capabilities can be elicited. We show this assumption does not hold under adversarial conditions by constructing backdoor attacks that embed encrypted neural networks within hos...
Andis Draguns, S. Motwani, Raymond Douglas et al.· 0 citations
Federated Large Language Models (FedLLMs) are promising to deliver privacy-preserving, collaborative intelligence. The decentralized training process is vulnerable to Byzantine attacks that corrupt semantic consistency and factual resilience through malicious model updates. This paper presents a new resilience strategy...
Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why solving difficult problems requires curiosity, judgment, agility—and an understanding of the rules. The post Learn the rules: The hacker mindset appeared first on Microsoft Research.
Critical flaws, compromised release channels and agentic AI threats landed within days of each other in mid-October 2026. Why “Black October” may say more about how well we now find weaknesses than about the attacks themselves. The post Black October? The Week AI Changed the Cybersecurity Conversation appeared first on GPT-Lab.
