相关论文: Peer-Preservation in Frontier Models

Federated learning (FL) enhances privacy by keeping user data on local devices. However, emerging attacks have demonstrated that the updates shared by users during training can reveal significant information about their data. This has…

Developers try to evaluate whether an AI system can be misused by adversaries before releasing it; for example, they might test whether a model enables cyberoffense, user manipulation, or bioterrorism. In this work, we show that…

Artificial intelligence (AI) models introduce privacy vulnerabilities to systems. These vulnerabilities may impact model owners or system users; they exist during model development, deployment, and inference phases, and threats can be…

This technical report presents methods developed by the UK AI Security Institute for assessing whether advanced AI systems reliably follow intended goals. Specifically, we evaluate whether frontier models sabotage safety research when…

Federated Inference (FI) studies how independently trained and privately owned models can collaborate at inference time without sharing data or model parameters. While recent work has explored secure and distributed inference from disparate…

As frontier AI models are deployed in high-stakes decision pipelines, their ability to maintain metacognitive stability (knowing what they do not know, detecting errors, seeking clarification) under adversarial pressure is a critical safety…

AI-based sensing at wireless edge devices has the potential to significantly enhance Artificial Intelligence (AI) applications, particularly for vision and perception tasks such as in autonomous driving and environmental monitoring. AI…

Differential privacy has become the gold standard for privacy-preserving machine learning systems. Unfortunately, subsequent work has primarily fixated on the privacy-utility tradeoff, leaving the subject of fairness constraints undervalued…

In this paper we study the resilience of peer-to-peer networks to preferential attacks. We define a network model and experiment with three di erent simple repairing algorithms, out of which the so called 2nd neighbor rewiring algorithm is…

Leading language model (LM) providers like OpenAI and Anthropic allow customers to fine-tune frontier LMs for specific use cases. To prevent abuse, these providers apply filters to block fine-tuning on overtly harmful data. In this setting,…

Automated control monitors could play an important role in overseeing highly capable AI agents that we do not fully trust. Prior work has explored control monitoring in simplified settings, but scaling monitoring to real-world deployments…

Federated learning (FL) faces a critical dilemma: existing protection mechanisms like differential privacy (DP) and homomorphic encryption (HE) enforce a rigid trade-off, forcing a choice between model utility and computational efficiency.…

Environmental understanding capability of $\textit{augmented}$ (AR) and $\textit{mixed reality}$ (MR) devices are continuously improving through advances in sensing, computer vision, and machine learning. Various AR/MR applications…

A core challenge for autonomous LLM agents in collaborative settings is balancing robust privacy understanding and preservation alongside task efficacy. Existing privacy benchmarks only focus on simplistic, single-turn interactions where…

Frontier AI systems are making transformative impacts across society, but such benefits are not without costs: models trained on web-scale datasets containing personal and private data raise profound concerns about data privacy and…

As artificial intelligence (AI) models are scaled up, new capabilities can emerge unintentionally and unpredictably, some of which might be dangerous. In response, dangerous capabilities evaluations have emerged as a new risk assessment…

This position paper argues that the AI/ML community should stop overclaiming and retire the label "positive backdoor," and instead treat trigger-activated hidden behaviors as Secret Alignment. Crucially, protective claims based on Secret…

Federated Learning (FL) enables collaborative model training while preserving data privacy; however, balancing privacy preservation (PP) and fairness poses significant challenges. In this paper, we present the first unified large-scale…

When language models are assigned professional personas, they face a conflict between maintaining the persona and disclosing their AI nature. How models resolve this conflict has practical consequences: a model that constructs detailed…

Instruction tuning has proven effective in enhancing Large Language Models' (LLMs) performance on downstream tasks. However, real-world fine-tuning faces inherent conflicts between model providers' intellectual property protection, clients'…