Maintaining the safety of large language models (LLMs) is crucial as they are increasingly deployed in real-world applications. Existing safety guardrails typically rely on single-pass classification or, more recently, distilled reasoning. Reasoning-based guardrails significantly outperform classification-only baseline...
Decompile-Diverge is proposed, a behavioral comparison oracle not relying on fixed or hand-crafted tests: for each function it synthesizes a driver, grows a fuzzing corpus from the reference, and reruns the decompiled code on the same inputs to detect changes in the function's behavior.
Chang Liu, Edward Raff, Kristopher K. Micinski· 3 citations
BUGSTONE-E2E, a framework that transforms vulnerability history into executable detection rules and validates their findings, demonstrates that CVE history can be turned into an executable workflow, transforming past vulnerabilities into reproducible detection and repair.
Qiu-Shi Wu, Kevin Eykholt, Youngja Park et al.· 0 citations
These results show that secure agent execution requires not only sound individual controls, but explicit contracts that preserve their guarantees across the complete instruction-to-effect path.
Chris Zheng, Yang Geng· 3 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
TIER, a Threat Implicitness Benchmark for behavioral safety evaluation of LLMs, is introduced, showing that safety behaviors evolve gradually across threat levels rather than shifting directly from refusal to compliance.
Thu-Hien Trinh-Thi, Hai-Yen Vong, Thanh-Ha Ung-Dung et al.· 0 citations
Fraudulent messages sent via Short Message Service (SMS) are increasingly obfuscated to evade cost-conscious classifiers in production systems. In Chinese SMS, attackers can exploit a wide range of carefully crafted obfuscation strategies to hide risk-bearing phrases while preserving human readability, making direct cl...
Jieyun Huang, Yi Shen, Kaikai Zhao et al.· 0 citations
Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache. Yet today's"forget"operations delete a plaintext memory record and stop, leaving every artifact derived from the revoked information intact. We f...
Chao Yao, Yang-Bo Wei, Zhen Huang et al.· 2 citations
Adaptive HMAS can provide accuracy cost tradeoff for ransomware analysis while retaining support for heterogeneous and incomplete modalities, as well as reducing average analysis cost and average analysis latency.
Prompt injection is widely recognized as a major security threat to AI agents that interact with untrusted external data, such as websites, documents, and emails. Prior work has shown that, in the text domain, black-box prompt injection can achieve near-perfect attack success rates (ASRs). In the image domain, however,...
Si-Zhe Chen, Yu-Lin Tsai, Ivan Evtimov et al.· 1 citation
This paper analyzes the implementation of blockchain-based integrity mechanisms in Greek Fiscal Electronic Mechanisms and the central tax information system “eSEND” and demonstrates that hardware-based fiscal mechanisms provide stronger guarantees for transaction completeness and tamper resistance than purely software-...
Panagiotis Mavridis, A. Baklezos, C. Nikolopoulos· 2026 7th International Confe...· 0 citations
This work identifies the attacker's adaptive search over the system attack surfaces as an important and underexplored security risk for tool-using agents and suggests that agentic security evaluations should characterize both the attacker's search procedure and compute budget.
D. M. Nguyen, Joon Sik Kim, Blazej Manczak et al.· 1 citation
This paper presents a meta-synthesis that draws together four constituent studies covering adversarial machine learning, AI-powered anomaly detection in cloud environments, automated vulnerability patching by multi-agent large language model (LLM) pipelines, and the broader landscape of securing AI systems across their...
Harsh Verma· International Journal of Sci...· 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.
