A novel message encoding paradigm is proposed called \textit{majority bit-aware encoding}, which relaxes the watermark signal strength from the green list size, which allows for a strong watermark signal to be preserved in generated texts even when using a large green list.
Jiahao Xu, Rui Hu, O. Kotevska et al.· arXiv.org· 3 citations
Differential privacy (DP) provides robust privacy guarantees for statistical inference, but this can lead to unreliable results and biases in downstream applications. While several noise-aware approaches have been proposed which integrate DP perturbation into the inference, they are limited to specific types of simple...
Talal Alrawajfeh, Joonas J\"alk\"o, Antti Honkela· 0 citations
A diagnostic-first methodology is proposed that localizes trigger-relevant features through constrained perturbation analysis and detects poisoning-induced behavioral drift by comparison with a distilled baseline model and enables targeted mitigation via automated data sanitization or neuron pruning.
Florian Spelter, J. Loevenich, Tobias Hürten et al.· 0 citations
Rising privacy concerns around large language models have accelerated research in private Transformer inference (PTI). Synthesizing the empirical claims in this line of research, however, is hindered by multi-dimensional cryptographic progress and inconsistent evaluation contracts. In this work, we systematize the fiel...
Model signing authenticates the full-precision weights W an LLM is released with, yet users run a quantized artifact Q whose producing transformation is not uniquely determined under current disclosure. We separate the non-environment inputs from the execution environment E and measure a quantity that, to our knowledge...
Finite safety assessments cannot exclude an adversary that preserves unattacked behavior until certification and activates afterward. We characterize this limitation for embodied and tool-use AI. An adversary that matches the complete observation law until observable, irrevocable certification and retains intervention...
Physical-layer spoofing attacks pose a critical threat to wireless localization systems by manipulating channel state information (CSI). Detecting such attacks becomes particularly challenging when the attacker and the legitimate node are spatially proximate, as their CSI characteristics become increasingly similar. To...
Conventional deepfake speech detectors primarily target synthesis artifacts and often degrade under unseen attacks or recording conditions. Human speech, however, contains physiological markers associated with respiration, phonation, and temporal coordination arising from the speech-production system that synthesizers...
Vojt\v{e}ch Stan\v{e}k, Michal Fri\v{c}, Kamil Malinka et al.· 0 citations
The semantic reconstruction game is introduced, a training-free evaluator that decomposes compliance judgments into constrained semantic comparisons and achieves the lowest mean error against human labels among evaluated methods and reduces variation in reported ASR.
The increasing adoption of differential privacy (DP) leads to public-facing DP deployments by both government agencies and companies. However, real-world DP deployments often do not fully disclose their privacy guarantees, which vary greatly between deployments. Failure to disclose certain DP parameters can lead to mis...
Onyinye Dibia, Mengyi Lu, Prianka Bhattacharjee et al.· 0 citations
Text watermarking has emerged as a crucial technique for identifying AI-generated content. However, current watermarking schemes entail a trade-off between text quality and watermark detectability, which is further intensified by the inherent low-entropy nature of code. To mitigate this dilemma, we propose Grammar-Driv...
As text-to-image (T2I) models advance and gain widespread adoption, their associated safety concerns are becoming increasingly critical. Malicious users exploit these models to generate Not-Safe-for-Work (NSFW) images using harmful or adversarial prompts, underscoring the need for effective safeguards to ensure the int...
Yiming Wang, Jiahao Chen, Qingming Li et al.· 0 citations
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.
