Category
1,407 papers
This paper proposes a new sign-recovery algorithm that achieves higher sign-recovery accuracy than the existing method, and enables efficient sign recovery even for trained models, and demonstrates end-to-end model extraction from models trained on MNIST and Fashion-MNIST.
This work develops a novel multi-draft speculative sampling algorithm based on Poisson processes that maintains both watermark strength and sampling efficiency, and is the first multi-draft, drafter-invariant speculative sampling scheme that maintains both watermark strength and sampling efficiency.
ServeGuard is a supply-chain primitive: the publisher ships a proof-carrying adapter whose proof lets a consumer or regulator verify, without the certified read factor and without trusting the publisher, that the adapter carries no hidden channel of this class relative to the declared monitor.
This work provides the first study of such a whole-system defense, especially with respect to a deployed and operational capability, and shows an increase in product abuse coverage, a 30% reduction in monthly alerts, and adaptability to changes in malicious actors'behavior.
Reach audiences
Advertise in front of researchers, engineers, and readers.
Reinforcement learning (RL) controllers have been recently adopted for Unmanned Aerial Vehicles (UAV) navigation and control. However, they are susceptible to action-space attacks that overwrite the action commands after the policy generates them and before the actuators execute them. While most existing defenses targe...
TERMon is a lightweight hardware runtime monitor that detects anomalies by observing inference behavior rather than re-executing or formally verifying the model, and represents class-conditional trusted behavior as hardware-efficient ternary patterns that are matched in parallel against a thermometer-encoded fingerprin...
CIPL (Channel Inversion for Privacy Leakage), a channel-aware evaluation framework for black-box privacy leakage in LLM agents, is presented and provides a common framework for comparing how internal sensitive dependence is realized as externally recoverable leakage across heterogeneous agent pipelines.
The results indicate that Micro-Collaborative Poisoning is not driven by a single dominant poisoned passage, but by the accumulation of weak adversarial signals across retrieved sources, which achieves downstream influence while leaving a weaker explicit poisoning signature than direct poisoning.
HE-Guardrail is proposed, a framework that evaluates guardrail mechanisms entirely over encrypted data and homomorphically controls whether the target-model response is returned to the client, with distinct security-efficiency-utility trade-offs.
This paper presents CESBench, 380 expert-written items across six sub-domains of cryptographic engineering security for IoT devices: side-channel, fault injection, implementation, countermeasures, evaluation, and integration, and presents the benchmark, prompts, and per-item results.
Under stated assumptions, it is proved post-cut issuer non-expansion, support-sound projection, compositional soundness under exact channel conservation, independent-support preservation, merge-order independence, and crash/replay stability.
DSTAR is proposed, a lightweight and efficient approach for distributed stochastic gradient descent that enhances robustness and convergence and outperforms other Byzantine-resilient methods that often suffer up to a 40-50\% accuracy drop under attack.
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.
GPT-Lab
· gpt-lab.eu
Oct 30, 2026
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.
Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities.
Gemini 3.8 Flash and 3.8 Flash Cyber deliver next-generation intelligence for agentic workflows and cybersecurity.
Google introduces Gemini 3.5 Flash Cyber, a lightweight cybersecurity model to find and patch vulnerabilities.
Students from the MIT Cybersecurity Clinic help local governments and other vulnerable organizations defend against digital threats.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
