Diverse Hardening Strategies

Follow this topic Ask about this topic

Momentum

2 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Jul 6Week of Sep 21

Latest papers 15

All topics

A Deterministic and Auditable AI Security Risk Assessment Framework with ATLAS Aligned Executable Rules and Formal Verification
Oct 1, 2026Yixuan Huang, Basel Halak, Boojoong KangArtificial Intelligence Risk Security Evaluation
University of Southampton, Southampton, UK
Skill2Env: Capability-Oriented Environment Synthesis from Skills for General Agents
Sep 27, 2026Weiyi Xu, Xiaowen Yang, Wen Da +7Synthetic Environments Skills
AllSpark Team
Bilinear Optimization Divergence: Diagnosing Factor-Constrained LoRA Continual Learning
Sep 20, 2026YongShun Wang, JianLin Su, Yong MaOrthogonality Bregman Divergences
Operating Multi-Node Full Fine-Tuning on NVIDIA B300: A Field Report on Telemetry-Based Triage, Negative Results, and Operational Hardening
Aug 6, 2026Seon Ho Kim, Ui Jeong Jeon, Su Hyeon Kim +1Nvidia Model Fine-Tuning
Samsung SDS
Distributing Security Controls Through Harness Engineering
Jul 28, 2026William Robert GoreSecurity Diverse Hardening Strategies
Georgia Institute of Technology
Beyond Fail-to-Pass: Iterative Hardening of Co-Generated Bug Reproduction Tests and Fixes
Jul 22, 2026Yuhao Tan, Zhibang Yang, Fangkai Yang +9Automated Program Repair Test Generation
1Nanjing University · 2Peking University · 3Microsoft
Data Leakage Prevention in Agentic Applications via Preemptive Hardening
Jul 21, 2026Akansha Shukla, Emily Bellov, Parth Atulbhai Gandhi +2Indirect Prompt Injection Diverse Hardening Strategies
Faculty of Computer and Information Science Ben-Gurion University of the Negev Beer-Sheva, Israel
i-EXAM: Instructable and Explainable Attack Connectivity Graph Modeler
Jul 7, 2026Rakesh Podder, Wadia Ganim, Sarath Sreedharan +2Security Evaluation Diverse Hardening Strategies
Colorado State University Fort Collins, Colorado, USA
Judging to Improve: A De-biased VLM-as-3D-Judge Protocol for Single-Image 3D Generation
Jun 18, 2026Ali Asaria, Tony Salomone, Deep GandhiLlm-As-A-Judge Protocol
Transformer Lab
Securing Multi-Agent GIS Systems: Risk Evaluation and Prompt Hardening Optimization
Jun 13, 2026Kyle Gao, Pranavi Kotta, Linlin Xu +2Diverse Hardening Strategies Agentic
Department of Systems Design Engineering, University of Waterloo, Waterloo, ON, Canada · Department of Mechatronics Engineering, University of Waterloo, Waterloo, ON, Canada · Department of Geomatics Engineering, University of Calgary, Calgary, AB, Canada +1
Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops
Jun 8, 2026Ziqian Zhong, Ivgeni Segal, Ivan Bercovich +3Diverse Hardening Strategies Agentic Benchmarks
Carnegie Mellon University · Fewshot Corp · Independent Researcher
SHIELDS: Automating OS Hardening with Iterative Multi-Agent Remediation
Jun 3, 2026Andrew Hamara, Dwight Horne, Aldehir Rojas +5Diverse Hardening Strategies Iterative
1L3Harris Technologies. · 2Texas A&M University.
Defense effectiveness across architectural layers: a mechanistic evaluation of persistent memory attacks on stateful LLM agents
May 8, 2026Jun Wen LeongLLM Defense Mechanisms Realistic Threat Model
Hardening x402: PII-Safe Agentic Payments via Pre-Execution Metadata Filtering
Apr 13, 2026Vladimir StantchevPayment Networks Metadata
SRH University Heidelberg, Germany · PRESIDIO Group, Sofia, Bulgaria
Hardening Soft Information: Evidence on Analyst Integration Costs
May 18, 2025Kerry Xiao, Amy ZangExacerbate Systemic Inefficiencies Diverse Hardening Strategies
Hong Kong University of Science and Technology

Oct 1, 2026, cs.AIJ/K move · Enter open · S save

A Deterministic and Auditable AI Security Risk Assessment Framework with ATLAS Aligned Executable Rules and Formal Verification

Yixuan Huang, Basel Halak, Boojoong Kang

University of Southampton, Southampton, UK

Artificial Intelligence Risk Security Evaluation Artificial Intelligence Evaluation Formal Verification Diverse Hardening Strategies Atlas+1

Open paper AskPDF arXiv

Artificial intelligence systems are increasingly deployed in high impact and safety critical settings, yet security assessment remains difficult to reproduce and defend under audit. Existing approaches often rely on narrative checklists or assessor driven scoring, and they lack an explicit, machine evaluable mapping from observable engineering artefacts to stable technique level outcomes. We present an evidence driven AI security assessment framework that operationalises assessment as a deterministic decision function. The framework normalises heterogeneous artefacts into a project independent Control ID taxonomy scored on a bounded four level ordinal scale, compiles technique level predicates from a pinned MITRE ATLAS snapshot via an explicit mitigation to control mapping, and outputs technique indexed feasibility and impact levels with traceable links back to the triggering evidence. We package all normative choices as a versioned assessment policy object to support repeatable reassessment across snapshots. To ensure semantic correctness, we formally verify boundedness, totality, ordered semantic consistency, and monotonicity of the compiled evaluator over the full declared score domain. We evaluate the framework on five public open source AI projects pinned to explicit repository snapshots, quantify before and after changes under a unified hardening intervention, and validate responsiveness to real engineering changes through fork based implementations of Software Bill of Materials (SBOM) generation and Continuous integration (CI) security scanning gates. Results show consistent downward shifts in feasibility profiles under strengthened observable controls, while worst case residual feasibility persists when technique specific core controls remain absent from the evidence scope.

Momentum

Related topics

Latest papers 15