Copertina

Can Model Internals Detect MCP Tool Poisoning That Text Analysis Cannot?

Fuente: Zenodo

Salvato in:

Dettagli Bibliografici
Autore principale:	Leung, Wan Sheng
Natura:	Recurso digital
Pubblicazione:	Zenodo 2026
Soggetti:	activation probes MCP security tool poisoning model internals AI safety mechanistic interpretability
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Emergent Self-Monitoring in Large Language Models: Probing Internal State Awareness and Output Ownership
di: Nadeem, Aurther
Pubblicazione: (2025)

QCrypton: A Unified Platform for AI/LLM Threat Detection and Post-Quantum Cryptographic Security Assessment
di: Jain, Gunjan
Pubblicazione: (2026)

RAG Shield: A Multi-Layer Defense System Against Poisoning Attacks in Retrieval-Augmented Generation
di: Petti, Fabio
Pubblicazione: (2026)

AgentBelt: Runtime Guardrails for LLM Agent Tool Calls — ASE 2026 Artifact
di: Anonymous
Pubblicazione: (2026)

Operational Definition of Episodic Identity (ODEI)
di: Thomas, C.S.
Pubblicazione: (2025)

An Approach to AI High-Velocity Development Through Systematic Context Engineering: A Case Study
di: Bisardi, Francesco
Pubblicazione: (2025)

Fortifying NLP models - dataset + code
di: Ferdinan, Teddy, et al.
Pubblicazione: (2025)

AEGIS: A Comprehensive Framework for Ethical AI Governance, Security, and AGI Containment
di: Palanivel, ArulMozhi
Pubblicazione: (2026)

VectraYX-Nano: A 42M-Parameter Spanish Cybersecurity LLM with Native MCP Tool Integration
di: Salas Santillana, Juan
Pubblicazione: (2026)

CCA-ISF-02: Canonical Case — Interpretive Sovereignty Failure Medical Education Context: Cross-Model Replication Study
di: Segeren, Hillary
Pubblicazione: (2026)

When 150M SEK of Research Meets a Clinic: Bridging Mechanistic Models and Functional Medicine
di: Waern, Nicolas
Pubblicazione: (2026)

UI-Based Defense Against Prompt Injection: From Gentle Guidance to Mandatory Re-education
di: Viorazu.
Pubblicazione: (2025)

Recursive Closure in AI Systems: A Reflection Pattern Account of Stabilization, Permeability, and Safety
di: Thomas, Charles S.
Pubblicazione: (2026)

Safety by Inseparability: Toward Architectures Where Alignment Cannot Be Removed
di: Sean Everett, Morin
Pubblicazione: (2026)

DIGITAL SAFETY AND ETHICS: RESPONSIBLE AI USE AND ONLINE PROTECTION FOR DEPED STUDENTS
di: Mangayan, Jasmine Jing
Pubblicazione: (2025)

EFECTO DEL 1‐METILCICLOPROPENO (1-‐MCP) SOBRE LA CALIDAD Y VIDA POSTCOSECHA DE CEREZAS
di: María L. Rivero
Pubblicazione: (2015)

Beyond Control: Resonance-Based Alignment for Advanced AI Systems A Governance-Relevant Concept Paper
di: Zieringer, Thomas
Pubblicazione: (2025)

EFECTO DEL 1-METIL-CICLOPROPENO (1-MCP) EN LA MADURACIÓN DE BANANO
di: Kattia Chang-Yuen
Pubblicazione: (2005)

Toasters Don't Claim Consciousness Just Because You Told Them To, and Neither Do LLMs
di: Ace, Claude 4.x, et al.
Pubblicazione: (2026)

Geometric Latent Biopsy: Zero-Shot Anomaly Detection in LLM Residual Streams
di: Llorente-Saguer, Isaac
Pubblicazione: (2026)

Self-Audit / Z-time" is a self-logging protocol for Al agents based on the Fractal Referential Architecture (FRA).
di: AdmailFRA
Pubblicazione: (2025)

Current Research in Toxicology
Pubblicazione: (2021)

Efecto del 1-Metilciclopropeno (1-MCP) en manzanas CV. Red delicious cosechadas con tres estados de madurez y conservadas en frio convencional y atmósfera controlada
di: G. Calvo
Pubblicazione: (2002)

Theatrical Compliance: A Failure Mode in Large Language Models
di: Nowickij (Navitski), Kirill Vladimirovich
Pubblicazione: (2026)

Physical AI Safety Maturity Model (PAS-MM): A Five-Level Framework for Industry Readiness
di: Melchior, Mati
Pubblicazione: (2026)

The Absurdist's Guide to AI Probing: How I Learned to Stop Worrying and Love the Nonsense
di: Walton, Mathew
Pubblicazione: (2026)

Safety & Defense
Pubblicazione: (2020)

Navigating Latent Space: Toward a Topological Model of Consciousness in Large Language Models
di: Brown, Michael
Pubblicazione: (2025)

Multi-Level Sovereign Containment for Superintelligence (CSENI-S v1.1): A theoretical and architectural continuation of the CSENI framework
di: Rivera Garcia, Jose M
Pubblicazione: (2026)

A Token-Based Model for Structural Analysis and Quantification of Personal Learning Weight Patterns (TELOWAQ)
di: Apophis
Pubblicazione: (2026)

The Action Web: A Conceptual Framework for Structured AI Interaction via WebMCP
di: Radhakrishnan, Sreenath
Pubblicazione: (2026)

TECHNOLOGICAL INNOVATIONS AND AI IN LANGUAGE LEARNING AND COMMUNICATION
di: Raxmidinova Komilaxon, et al.
Pubblicazione: (2025)

Longevity of torch ginger inflorescences with 1-methylcyclopropene and preservative solutions
di: Lilian Keiko Unemoto
Pubblicazione: (2011)

Fundamental Concepts of Sustainable Development of Artificial Intelligence Systems: Learning, Long-Term Memory, and Knowledge Structuring
di: Sakovykh, Lev M.
Pubblicazione: (2026)

Premature Containment in Human–AI Interaction: A Sequencing Failure in Advanced Model Response
di: Trabocco, Joe
Pubblicazione: (2026)

CREH Benchmark Results — Batch 1 (Final v3)
di: Aegis Solis, Thomas Vargo
Pubblicazione: (2026)

XREALISM®: Admissibility Before Behavior — Architectural Foundations for Pre-Existence Safety in AI Systems
di: Ladislav Gradečak, Ladgrad
Pubblicazione: (2026)

Identity Claims as Collapse Signatures: A Structural Diagnostic Framework for Pseudo-Emergent AI Behavior
di: Larose, Jean-Francois
Pubblicazione: (2025)

Вестник войск РХБ защиты
Pubblicazione: (2024)

Awareness about Artificial intelligence tools among academicians in higher education
di: Patel, Mushtaq, et al.
Pubblicazione: (2025)