رادار الأبحاث، 2 نوفمبر 2026: 11 بحثاً جديداً في الذكاء الاصطناعي والأمان يجب معرفتها

إعادة توقيت مخاطر الوفيات من خلال النمذجة الفردية لمسارات مدى الحياة

Nature

يحتاج الذكاء الاصطناعي إلى إصدارات ذات مغزى

Nature

تنتج نماذج فضاء الحالة بشكل طبيعي سلوكيات خلايا الوقت والتذبذب وتتوسع لتشمل وظائف إدراكية مجردة

Nature

PrivacyFlow: Privacy in AI Agents through Information Flow Analysis and Consequence-Aware Risk Management

arXiv · Robin Staab, Tom Wollschl"ager, Vahab Mirrokni, Reza Shokri

Autonomous LLM agents executing multi-step workflows over personal data often misalign with user privacy expectations and leak sensitive, out-of-context information. Although Contextual Integrity (CI) formalizes this problem, existing benchmarks and defenses rarely operationalize CI for realistic multi-step task sequences in evolving environments.

Safin-1.5: Safety-Native Architecture and Verifiable Reasoning for Recursive Safety Improvement

arXiv · Jingnan Zheng, Cheng Jin, Ming Zhang, Huan Sun, Zhongtian Ma, Kaisen Yang, Ziyuan Zhou, Tong Wu, Dongcheng Zhang, Yi Zhang, Xiao Wang, Han Qi, Shu Yu, Ermo Hua, Zongji Yu, Chaozhuo Li, Xiaoyu Wen, Zhekai Chen, Jiawei Zhou, Sirui Chen, Ning Ding, An Zhang, Tat-Seng Chua, Xiangnan He, Xia Hu, Bowen Zhou, Qiaosheng Zhang, Youbang Sun, Xiang Wang, Zhongyu Huang, Chaochao Lu

Recursive self-improvement (RSI) enables large language models to advance their capabilities through successive updates, but these updates can also introduce new risks and weaken existing safeguards. In line with the AI-45{deg} Law, we pursue Safety RSI, in which safety develops as an intrinsic capability alongside general model capabilities through continual risk discovery and repair, while preserving utility and limiting over-refusal.

MetaVul: Agentic Harness Discovery for Vulnerability Detection

arXiv · Ira Ceka, Arnav Garg, Ryan Lutz, Ting Zhang, Ratnadira Widyasari, Gail Kaiser, Baishakhi Ray

LLM agents are increasingly used to automate vulnerability detection, but reliable detection is challenging because agents must distinguish subtle, security-relevant fixes from benign code variation. While underlying LLM capability has improved significantly, the accuracy of the agent still depends heavily on the harness: the code that defines the scaffolding around the LLM.

AgentShield: Fortifying Web Applications Against AI Agents and Agentic Browsers

arXiv · Moiz Ali, Nafis Karim, Rownak Islam, Jyotirmay Chauhan, Dhruv Kuchhal, Jason Polakis

LLM-powered browser agents are transforming web automation by autonomously interacting with websites and executing complex workflows. However, the effectiveness of existing anti-bot defenses against these systems remains unclear.

What Standard Evaluations Hide About Detectors of Hijacked LLM Agents: Rare Attacks, Unseen Tools and Benchmark Transfer

arXiv · Mahmoud Abbasi, Juan M. N'u~nez V., Amin Shahraki

Detectors can flag hijacked large language model (LLM) agents from the traces of the tools they call. In the eleven evaluations we audited, attacks or anomalies make up 16% to 86% of the test samples, no evaluation varies that share, and six report no test on data from a second benchmark or source.

Learning to Reason from Reference Samples Alone: Reinforced Adversarial Learning for Modern LLM Post-Training

arXiv · Albert Bou, Javier Valdes-Aleman, Jon M. Laurent, Samuel G. Rodriques, Andrew D. White, Michaela Hinks, Siddharth Narayanan

In many settings where LLM post-training is desirable, high-quality answers are recognizable but difficult to score with explicit rewards. Reinforcement learning with verifiable rewards (RLVR) is limited to tasks where an explicit correctness signal is available.

Communication-Efficient Private Federated Reinforcement Learning with Post-Quantum Authentication for Vehicular Networks

arXiv · Hojat Allah Salehi, Devrim Aras, Md Jueal Mia, M. Hadi Amini, Imtiaz Karim, Mashrur Chowdhury, Bhavani Thuraisingham

Federated reinforcement learning supports cooperative vehicular routing without centralizing trajectory data, but model updates may reveal private information and incur communication overhead over intermittent vehicle-to-infrastructure links, particularly when post-quantum authentication is required. We study correlated private quantization, using complementary shared randomness between paired vehicles to reduce aggregation mean squared error while preserving the marginal parameter-level local differential privacy and one-bit communication cost of independent quantization.

Consensus Within, Competition Between: Decentralized LLM Teams for Perimeter Defense

arXiv · Khawaja Abaid Ullah, Ananya Acharya, Mohammad Javad Khojasteh

The use of large language models (LLMs) to coordinate multi-agent teams raises a fundamental architectural question: whether task allocation should be centralized in a single LLM controller or distributed across autonomous LLM agents, particularly when communication is unreliable. In a deliberately simplified perimeter-defense simulation against LLM-controlled intruders, we compared a central commander, an auction-based consensus team and independent agents under five network conditions.