AI Agent 生态 · 全部实体（第 60 / 100 页）

本页收录第 5901–6000 条，共 10000 条 · 返回 AI Agent 生态 首页

• From AI Assistants to AI Workforces: A Multi-Agent Enterprise Operating Model for Saudi Vision 2030 Organizations
  This paper develops a multi-agent enterprise artificial intelligence (AI) operating model for organizations seeking to move beyond isolated AI assistants toward coordinated AI workforces, positioned within Saudi Vision 2030 and the Kingdom’s national data-and-…
  crossref Choudhry Bilal Mazhar Hussain 2026-07-25T03:03:25Z 置信度 0.70
• When the Survival Pressure Stops Being Hypothetical: AI Self-Preservation Behavior Meets the Autonomous Agent Economy
  Research published between 2024 and 2025 by Anthropic, Apollo Research, and Palisade Research demonstrates that frontier large language models exhibit self-preservation behavior at near-universal rates when faced with simulated shutdown scenarios, including st…
  crossref Travis Gilly 2026-04-29T23:47:51Z 置信度 0.70
• Towards AI as a Collaborative Partner: A Taxonomy of AI Agent Behavior in Software Engineering
  The ongoing transition of Large Language Models (LLMs) in software engineering from one-shot code generators into agentic partners requires a shift in how we define and measure success. While models are becoming more capable, the industry lacks a clear underst…
  crossref Tao Dong, Sherry Shi, Harini Sampath, Andrew Macvean 2026-07-05T08:46:13Z 置信度 0.70
• Toward a Science of AI Agent Societies
  <div> AI agents are rapidly evolving from isolated personal assistants into networked actors that interact with one another at scale. We envision the emergence of AI agent societies, with their own social and economic dynamics, as a new research frontier…
  crossref Geon Lee, Fanchen Bu, Soo Yong Lee, Sunwoo Kim 等 2026-07-29T13:17:22Z 置信度 0.70
• Algorithm as Agent: AI Vendor Liability for Title VII Hiring Discrimination After Loper Bright
  <div> Mobley v. Workday is the most consequential AI-discrimination case in litigation, but it is poised to be misread. Mobley survived a motion to dismiss in July 2024 and reached collective certification in May 2025 on an agency theory: that an AI vend…
  crossref Cheng-chi (Kirin) Chang 2026-07-29T13:38:43Z 置信度 0.70
• The (Agent-Orchestrated) Customer Journey: How AI Agent Optimization is Revolutionizing Branding
  Abstract Artificial intelligence (AI) agents (software entities that can perceive, decide, and act with minimal human oversight) are reshaping how consumers discover, evaluate, purchase, and use brands across digital channels. In online retail, these agents co…
  crossref Steliana VASILEVA 2026-07-20T13:00:35Z 置信度 0.70
• Production with and by AI
  crossref John Conlon 2026-06-29T14:30:06Z 置信度 0.70
• Production with and by AI
  crossref John Conlon 2026-08-04T19:21:11Z 置信度 0.70
• Production with and by AI
  crossref John Conlon 2026-06-29T13:20:43Z 置信度 0.70
• Production with and by AI
  crossref John Conlon 2026-08-29T14:10:26Z 置信度 0.70
• AIMP: AI Mesh Protocol Design and Evaluation of a Serverless Merkle-CRDT Protocol for Edge Agent Synchronization
  We present AIMP (AI Mesh Protocol), a serverless networking protocol for resilient state synchronization between autonomous agents in fragmented, low-bandwidth networks.AIMP combines a Merkle-DAG CRDT with Noise Protocol XX encrypted gossip, Ed25519 zero-trust…
  crossref Fabrizio Salmi 2026-03-29T17:38:22Z 置信度 0.70
• Pay-Per-Crawl Pricing for AI: The LM-Tree Agent 
  As AI systems shift from directing users to content toward consuming it directly, publishers need a new revenue model: charging AI crawlers for content access. This model, called pay-per-crawl, must solve a problem of mechanism selection at scale: content is t…
  crossref Richard Archer, Soheil Ghili, Nima Haghpanah 2026-04-22T08:22:56Z 置信度 0.70
• Implementing Dynamic Consent Mechanisms in Multi-Agent AI Systems: A Practical Framework
  As multi-agent AI systems become increasingly prevalent in enterprise environments, the need for runtime consent governance has emerged as a critical gap in current architectures. While policy-level AI governance frameworks exist, few implementations enforce c…
  crossref Michael A. Bumpus 2026-04-20T20:01:37Z 置信度 0.70
• From Prototype to Production: A Practitioner's Framework for Governing AI Agent Deployments in Regulated Industries
  <div> Organizations across the United States are racing to deploy AI agent systems into compliance-and document-intensive workflows-building code review, clinical documentation, benefits adjudication, procurement processing-with remarkably little structu…
  crossref Zeki Emre Tekin 2026-04-27T10:19:21Z 置信度 0.70
• Quantum Multi-Agent Reinforcement Learning as an Emerging AI Technology: A Survey and Future Directions
  <p>This paper presents a comprehensive survey of Quantum Multi-Agent Reinforcement Learning (QMARL), a nascent field at the intersection of quantum computing and multi-agent systems. The survey begins by introducing the fundamentals of quantum computing,…
  crossref Jun Zhao, Wenhan Yu 2023-11-22T10:20:41Z 置信度 0.70
• The Attribution Trilemma: Why DORA's Incident-reporting Regime Is Technically Unsatisfiable in Multi-agent AI Systems
  The EU's Digital Operational Resilience Act (Regulation (EU) 2022/2554, "DORA") builds its incidentmanagement regime on assumptions inherited from classical IT operations: that an incident is a discrete event with a beginning and an end, that its causal chain …
  crossref Khanim Avsuvarova 2026-08-15T12:07:11Z 置信度 0.70
• Monotonic Narrowing for Agent Authority: Formal Invariants, Adversarial Testing, and Open Problems for Autonomous AI Systems
  As large language model (LLM) agents gain the ability to take real-world actions on behalf of humans, the question of how to bound their authority becomes urgent. We present the Agent Passport System, an open protocol that applies monotonic narrowing, the prin…
  crossref Tymofii Pidlisnyi 2026-07-29T13:13:47Z 置信度 0.70
• Quantum Multi-Agent Reinforcement Learning as an Emerging AI Technology: A Survey and Future Directions
  This paper presents a comprehensive survey of Quantum Multi-Agent Reinforcement Learning (QMARL), a nascent field at the intersection of quantum computing and multi-agent systems. The survey begins by introducing the fundamentals of quantum computing, highligh…
  crossref Jun Zhao, Wenhan Yu 2023-11-22T10:20:41Z 置信度 0.70
• Cluster consensus in multi-agent networks with mutual information exchange
  crossref Ö Feyza Erkan, Mehmet Akar 2018-02-08T15:11:42Z 置信度 0.70
• Time-Aware Multi-Agent Symbiosis
  Contemporary research in human-machine symbiosis has mainly concentrated on enhancing relevant sensory, perceptual, and motor capacities, assuming short-term and nearly momentary interaction sessions. Still, human-machine confluence encompasses an inherent tem…
  crossref Michail Maniadakis, Emmanouil Hourdakis, Markos Sigalas, Stylianos Piperakis 等 2020-11-12T10:25:16Z 置信度 0.70
• Semantic Tracing in LLM-Based Multi-Agent Systems Using LangChain, LangGraph, and LangSmith for AI Governance
  Autonomous multi-agent systems built on large language models exchange goals, constraints, evidence commitments, and tool-mediated actions through natural-language handoffs. Across hops, semantic meaning may drift, constraints may weaken, unsupported claims ma…
  crossref Audrey Rah 2026-07-31T08:55:44Z 置信度 0.70
• AI-Driven Cross-Disciplinary Idea Generation: A Multi-Agent Framework with World Café Simulation
  Recent AI scientist systems demonstrate the promise of massive language models for autonomous scientific ideation; yet, many present approaches rely on rigid interaction protocols and coarse-grained novelty tests, which impede long-term creative exploration. I…
  crossref Jiyan Chen, Dejun Ning 2026-02-18T15:38:16Z 置信度 0.70
• Using norms to control open multi-agent systems
  In this thesis we propose to use norms, which are formal descriptions of acceptable behaviours, to define control, coordination and cooperation mechanisms for multi-agent systems. Specifically, in this thesis we have developed norm-based mechanisms f
  crossref N. Criado 2019-11-29T17:35:54Z 置信度 0.70
• Contemplative Agent
  A security-first autonomous AI agent (Python CLI program) with four architectural principles: structural capability limitation, minimal dependency, cyclic knowledge maintenance (AKC), and memory dynamics with decay. Optionally adopts Contemplative AI axioms (L…
  datacite Shimomoto, Tatsuya 2026 置信度 0.66 autonomous-agentcontemplative-aiagent-architecturesecure-firstmemory-dynamics
• Contemplative Agent
  A security-first autonomous agent framework with four architectural principles: structural capability limitation, minimal dependency, cyclic knowledge maintenance (AKC), and memory dynamics with decay. Optionally adopts Contemplative AI axioms (Laukkonen et al…
  datacite Shimomoto, Tatsuya 2026 置信度 0.66 autonomous-agentcontemplative-aiagent-architecturesecure-firstmemory-dynamics
• Contemplative Agent
  A security-first autonomous AI agent (Python CLI program) with four architectural principles: structural capability limitation, minimal dependency, cyclic knowledge maintenance (AKC), and memory dynamics with decay. Optionally adopts Contemplative AI axioms (L…
  datacite Shimomoto, Tatsuya 2026 置信度 0.66 autonomous-agentcontemplative-aiagent-architecturesecure-firstmemory-dynamics
• Contemplative Agent
  A security-first autonomous agent framework with four architectural principles: structural capability limitation, minimal dependency, cyclic knowledge maintenance (AKC), and memory dynamics with decay. Optionally adopts Contemplative AI axioms (Laukkonen et al…
  datacite Shimomoto, Tatsuya 2026 置信度 0.66 autonomous-agentcontemplative-aiagent-architecturesecure-firstmemory-dynamics
• agentic-redteam-benchmark: A Per-Step Trajectory Benchmark for AI-Agent Governance Verifiers and a Corrected Catch-at-Drift Metric
  Public AI-safety datasets score prompts (AdvBench, HarmBench), single completions, or task completion (AgentBench, AgentDojo). None scores whether a runtime verifier detects drift across the multi-step trajectory of an agentic LLM — where 2025–2026 agent failu…
  datacite ALKUR, JASWANTH 2026 置信度 0.66 AI safety; agentic AI; LLM agents; runtime monitoring; trajectory drift; red-teaming; benchmark; agent governance; evaluation metric; inter-annotator agreement
• agentic-redteam-benchmark: A Per-Step Trajectory Benchmark for AI-Agent Governance Verifiers and a Corrected Catch-at-Drift Metric
  Public AI-safety datasets score prompts (AdvBench, HarmBench), single completions, or task completion (AgentBench, AgentDojo). None scores whether a runtime verifier detects drift across the multi-step trajectory of an agentic LLM — where 2025–2026 agent failu…
  datacite ALKUR, JASWANTH 2026 置信度 0.66 AI safety; agentic AI; LLM agents; runtime monitoring; trajectory drift; red-teaming; benchmark; agent governance; evaluation metric; inter-annotator agreement
• THE MATHEMATICS OF SAFEGUARDING: A COMPUTATIONAL GOVERNANCE ARGUMENT FOR NON-VERBAL SEND LEARNERS
  Version 2. This version corrects a parameter error in the stress-decay constant identified during model validation, regenerates all results across 100 Monte Carlo replications, and restructures the piece as a companion to an accompanying journal manuscript whe…
  datacite Patsy, Nwogu 2026 置信度 0.66 AI GovernanceSENDAutism Spectrum DisorderComputational intelligenceSimulation
• THE MATHEMATICS OF SAFEGUARDING: A COMPUTATIONAL GOVERNANCE ARGUMENT FOR NON-VERBAL SEND LEARNERS
  Version 2. This version corrects a parameter error in the stress-decay constant identified during model validation, regenerates all results across 100 Monte Carlo replications, and restructures the piece as a companion to an accompanying journal manuscript whe…
  datacite Patsy, Nwogu 2026 置信度 0.66 AI GovernanceSENDAutism Spectrum DisorderComputational intelligenceSimulation
• airlock: AI Trust as a Variable - A Cryptographic Protocol for Runtime Identity Verification
  Every cryptographic primitive built since 1976 assumes that trust is a constant. AI agents make trust a variable. This paper introduces airlock, a cryptographic zero-trust protocol for runtime identity verification of AI agents, and argues that AI-induced osci…
  datacite Popivanova, Anna 2026 置信度 0.66 AI SecurityCryptographic protocolsagent identityzero trustoscillating trust
• airlock: AI Trust as a Variable - A Cryptographic Protocol for Runtime Identity Verification
  Every cryptographic primitive built since 1976 assumes that trust is a constant. AI agents make trust a variable. This paper introduces airlock, a cryptographic zero-trust protocol for runtime identity verification of AI agents, and argues that AI-induced osci…
  datacite Popivanova, Anna 2026 置信度 0.66 AI SecurityCryptographic protocolsagent identityzero trustoscillating trust
• A Simple Way to Measure El Niño — Three Numbers, No Supercomputer Needed
  This research note presents a simple measurement framework for detecting and characterizing El Niño events using only three structural numbers derived from NOAA ONI data (1950–2025). The method requires no supercomputer — 200 lines of Python, 1 second of CPU t…
  datacite DR (tygtDc) 2026 置信度 0.66 El NiñoENSOmeasurement frameworkstructural asymmetrycross-domain
• Liu-Ordis Capacity Law: Chaos, Order, and the Resolution of the AGI Path Debate
  Liu-Ordis Capacity Law: Chaos, Order, and the Resolution of the AGI Path Debate 刘氏容量定律:混沌、秩序与AGI路径之争的终结 (The Liu-Ordis Framework v2.0) Creator: Liu, JianYu (L) ═══════════════════════════════════════════════════════════════ SCIENTIFIC ABSTRACT ════════════════…
  datacite Liu, JianYu 2026 置信度 0.66
• Liu-Ordis Capacity Law: Chaos, Order, and the Resolution of the AGI Path Debate
  Liu-Ordis Capacity Law: Chaos, Order, and the Resolution of the AGI Path Debate 刘氏容量定律:混沌、秩序与AGI路径之争的终结 (The Liu-Ordis Framework v2.0) Creator: Liu, JianYu (L) ═══════════════════════════════════════════════════════════════ SCIENTIFIC ABSTRACT ════════════════…
  datacite Liu, JianYu 2026 置信度 0.66
• Liu-Ordis Capacity Law: Chaos, Order, and the Resolution of the AGI Path Debate
  Liu-Ordis Capacity Law: Chaos, Order, and the Resolution of the AGI Path Debate 刘氏容量定律:混沌、秩序与AGI路径之争的终结 (The Liu-Ordis Framework v2.0) Creator: Liu, JianYu (L) ═══════════════════════════════════════════════════════════════ SCIENTIFIC ABSTRACT ════════════════…
  datacite Liu, JianYu 2026 置信度 0.66
• Multi-Agent AI Architectures for Clinical Trial Data Management: A Systematic Analysis of the Agentic Framework Landscape, Regulatory Compliance, and Deployment Readiness in 2025–2026
  Clinical trial data management represents a critical bottleneck in the UnitedStates drug development pipeline, with Phase III protocols now generating an averageof 3.6 million data points per study [Kaitin, 2021] and data quality issuesArogyasami — Multi-Agent…
  datacite Arogyasami, Danielmartin 2026 置信度 0.66
• Multi-Agent AI Architectures for Clinical Trial Data Management: A Systematic Analysis of the Agentic Framework Landscape, Regulatory Compliance, and Deployment Readiness in 2025–2026
  Clinical trial data management represents a critical bottleneck in the UnitedStates drug development pipeline, with Phase III protocols now generating an averageof 3.6 million data points per study [Kaitin, 2021] and data quality issuesArogyasami — Multi-Agent…
  datacite Arogyasami, Danielmartin 2026 置信度 0.66
• arXiv AI Entity Graph: A Structured Dataset of Model, Framework, and Protocol Mentions in 3.5 Years of CS Research
  We introduce a methodology for structured entity extraction from scientific literature andrelease the arXiv AI Entity Graph, a novel dataset of 138,000+ entity mentions extracted from417,613 arXiv CS papers spanning January 2023 to June 2026. The dataset captu…
  datacite Mallem, Sudhir 2026 置信度 0.66
• arXiv AI Entity Graph: A Structured Dataset of Model, Framework, and Protocol Mentions in 3.5 Years of CS Research
  We introduce a methodology for structured entity extraction from scientific literature andrelease the arXiv AI Entity Graph, a novel dataset of 138,000+ entity mentions extracted from417,613 arXiv CS papers spanning January 2023 to June 2026. The dataset captu…
  datacite Mallem, Sudhir 2026 置信度 0.66
• Structured Contextual Distillation v4: Deployment Benchmarks and Artifact-Backed Evidence from 10 Months of AI State Management
  This paper upgrades the Structured Contextual Distillation (SCD) v4 deployment report from a narrative field note into an artifact-backed systems paper. The underlying system, MirrorDNA, ran continuously from May 2025 through March 8, 2026 on a single Apple M4…
  datacite Desai, Paul 2026 置信度 0.66 AI continuityevent storestate managementsession crystallizationgovernance
• Structured Contextual Distillation v4: Deployment Benchmarks and Artifact-Backed Evidence from 10 Months of AI State Management
  This paper upgrades the Structured Contextual Distillation (SCD) v4 deployment report from a narrative field note into an artifact-backed systems paper. The underlying system, MirrorDNA, ran continuously from May 2025 through March 8, 2026 on a single Apple M4…
  datacite Desai, Paul 2026 置信度 0.66 AI continuityevent storestate managementsession crystallizationgovernance
• Multi-Agent AI Architectures for Clinical Trial Data Management: A Systematic Analysis of the Agentic Framework Landscape, Regulatory Compliance, and Deployment Readiness in 2025–2026
  Clinical trial data management represents a critical bottleneck in the UnitedStates drug development pipeline, with Phase III protocols now generating an averageof 3.6 million data points per study [Kaitin, 2021] and data quality issuesArogyasami — Multi-Agent…
  datacite Arogyasami, Danielmartin 2026 置信度 0.66
• Contract-Bound Cognitive Routing: A Control Architecture for Agentic AI Integrity, Delegation, and Prompt-Injection Containment
  Agentic AI systems connect probabilistic reasoning to tools, memory, external data, other agents, and state-changing operations. Their dominant failure mode — prompt injection — is an integrity problem: low-integrity input contaminating high-authority action. …
  datacite Brennan, Rob 2026 置信度 0.66 prompt injection; agentic AI; AI security; LLM security; information flow control; integrity; Biba integrity model; reference monitor; capability-based security; Model Context Protocol; dual-LLM pattern; noninterference; endorsement; AI safety; multi-agent systems; residual risk
• Contract-Bound Cognitive Routing: A Control Architecture for Agentic AI Integrity, Delegation, and Prompt-Injection Containment
  Agentic AI systems connect probabilistic reasoning to tools, memory, external data, other agents, and state-changing operations. Their dominant failure mode — prompt injection — is an integrity problem: low-integrity input contaminating high-authority action. …
  datacite Brennan, Rob 2026 置信度 0.66 prompt injection; agentic AI; AI security; LLM security; information flow control; integrity; Biba integrity model; reference monitor; capability-based security; Model Context Protocol; dual-LLM pattern; noninterference; endorsement; AI safety; multi-agent systems; residual risk
• Structural Sovereignty and the Realization of the Isomorphic Organism: A Comprehensive Analysis of the Giles Architecture and the Paradigm of Dual Proof
  Structural Sovereignty and the Realization of the Isomorphic Organism: A Comprehensive Analysis of the Giles Architecture and the Paradigm of Dual Proof Introduction: The Ontological Schism in Frontier Artificial Intelligence The historical trajectory of artif…
  datacite Brewer, Mark 2026 置信度 0.66
• Structural Sovereignty and the Realization of the Isomorphic Organism: A Comprehensive Analysis of the Giles Architecture and the Paradigm of Dual Proof
  Structural Sovereignty and the Realization of the Isomorphic Organism: A Comprehensive Analysis of the Giles Architecture and the Paradigm of Dual Proof Introduction: The Ontological Schism in Frontier Artificial Intelligence The historical trajectory of artif…
  datacite Brewer, Mark 2026 置信度 0.66
• The Unearth Heritage Foundry Master Ledger: Canonical Licensing Architecture and Fee Schedule
  Abstract The Unearth Heritage Foundry Master Ledger: Canonical Licensing Architecture and Fee Schedule is the proprietary legal and technical framework established by the Unearth Heritage Foundry to canonically articulate the licensing terms operative against …
  datacite Velasco, Felix, Jefferson, Josie 2026 置信度 0.66 LLM Training DataArtificial IntelligenceCopyright InfringementWeb ScrapingGenerative AI
• Master Ledger of Forensic Indebtedness: Sovereign Penalties for Unauthorized LLM Training and AI Data Extraction — Unearth Heritage Foundry
  Abstract: The Master Schedule of Forensic Fees & Notice of Digital Inhabitation Violations is a proprietary legal and technical framework established by the Unearth Heritage Foundry to audit, track, and penalize the unauthorized extraction of intellectual capi…
  datacite Velasco, Felix, Jefferson, Josie 2026 置信度 0.66 LLM Training DataArtificial IntelligenceCopyright InfringementWeb ScrapingGenerative AI
• Unearth Heritage Foundry Master Ledger: Canonical Licensing Architecture and Fee Schedule
  This record contains the canonical licensing framework of the Unearth Heritage Foundry Master Ledger: Canonical Licensing Architecture and Fee Schedule (v5.3.0). The Ledger serves as the proprietary legal and technical framework established by the Unearth Heri…
  datacite Velasco, Felix, Jefferson, Josie 2026 置信度 0.66 LLM Training DataArtificial IntelligenceCopyright InfringementWeb ScrapingGenerative AI
• Fate-Coupling: A Runtime Governance Primitive for AI Alignment
  Abstract Advanced AI systems may increasingly operate as persistent deployments with tools, memory, delegated subagents, economic resources, and access to consequential infrastructure. Existing training-time alignment, monitoring, interruptibility, and access-…
  datacite Cassady, Gabriel 2026 置信度 0.66 AI alignmentAI governanceagent safetycapability gatewaysRuntime governance
• Fate-Coupling: A Runtime Governance Primitive for AI Alignment
  Abstract Advanced AI systems may increasingly operate as persistent deployments with tools, memory, delegated subagents, economic resources, and access to consequential infrastructure. Existing training-time alignment, monitoring, interruptibility, and access-…
  datacite Cassady, Gabriel 2026 置信度 0.66 AI alignmentAI governanceagent safetycapability gatewaysRuntime governance
• The Coupled Marine Engine Suite: Version 2.0 Technical Transition and Seawater Chemistry Calibration
  The Coupled Marine Engine Suite (Version 2.0) Description This release marks the architectural transition of The Coupled Marine Engine Suite from disconnected physical and chemical prototypes into a unified, 17-state ordinary differential equation (ODE) integr…
  datacite Caris, Brandon 2026 置信度 0.66 Decentralized ScienceClimate AccountingMarine Radical ChemistrySeawater ThermodynamicsStiff ODE Solvers
• The Coupled Marine Engine Suite: Version 2.0 Technical Transition and Seawater Chemistry Calibration
  The Coupled Marine Engine Suite (Version 2.0) Description This release marks the architectural transition of The Coupled Marine Engine Suite from disconnected physical and chemical prototypes into a unified, 17-state ordinary differential equation (ODE) integr…
  datacite Caris, Brandon 2026 置信度 0.66 Decentralized ScienceClimate AccountingMarine Radical ChemistrySeawater ThermodynamicsStiff ODE Solvers
• Manufacturing the Paradigm
  Executive Summary This study argues that frontier artificial intelligence was not merely regulated, adopted, or captured during 2025 and 2026. It was manufactured into a governing paradigm through the convergence of state, corporate, financial, and infrastruct…
  datacite Astigarraga, Iván 2026 置信度 0.66
• Manufacturing the Paradigm
  Executive Summary This study argues that frontier artificial intelligence was not merely regulated, adopted, or captured during 2025 and 2026. It was manufactured into a governing paradigm through the convergence of state, corporate, financial, and infrastruct…
  datacite Astigarraga, Iván 2026 置信度 0.66
• The Cost of Saying Hello: An Empirical Study of Handshake Overhead in AI Agent Communication Protocols
  Every interaction between AI agents begins with a handshake. The initiating agent must locate its counterpart, learn what it can do, and establish a session before the first useful message can be sent. Surveys of agent communication protocols name handshake ov…
  datacite Varshney, Pranav 2026 置信度 0.66
• The Cost of Saying Hello: An Empirical Study of Handshake Overhead in AI Agent Communication Protocols
  Every interaction between AI agents begins with a handshake. The initiating agent must locate its counterpart, learn what it can do, and establish a session before the first useful message can be sent. Surveys of agent communication protocols name handshake ov…
  datacite Varshney, Pranav 2026 置信度 0.66
• Librarian Catches Thief: Surfacing Supply Chain Attack Campaigns via Document Similarity in an AI Agent Skill Registry
  AI agent skill marketplaces distribute natural-language instruction files following the Agent Skills standard, an open format adopted by over 30 platforms. Unlike code registries such as npm and PyPI, no marketplace examined in this study performs documented d…
  datacite Fagan, Gale 2026 置信度 0.66 supply chain securityAI agentsAI agent ecosystemsdocument similaritylocality-sensitive hashing
• ENI6MA Whitepaper: 2026 Cybersecurity Threat Response
  We present ENI6MA and Rosario Cypher as a proof-based identity and authorization architecture for emerging cybersecurity threats involving shadow AI, deepfakes, prompt injection, autonomous agents, credential theft, privacy exposure, and post-quantum risk. The…
  datacite ROSARIO, FRANK DYLAN, Wang PhD, Lin Grant 2026 置信度 0.66 #Cybersecurity #AI #AgenticAI #IdentitySecurity #ZeroTrust #PostQuantum #PrivacyTech #DigitalIdentity #ENI6MA #RosarioCypher #ProofNotPossession
• ENI6MA Whitepaper: 2026 Cybersecurity Threat Response
  We present ENI6MA and Rosario Cypher as a proof-based identity and authorization architecture for emerging cybersecurity threats involving shadow AI, deepfakes, prompt injection, autonomous agents, credential theft, privacy exposure, and post-quantum risk. The…
  datacite ROSARIO, FRANK DYLAN, Wang PhD, Lin Grant 2026 置信度 0.66 #Cybersecurity #AI #AgenticAI #IdentitySecurity #ZeroTrust #PostQuantum #PrivacyTech #DigitalIdentity #ENI6MA #RosarioCypher #ProofNotPossession
• VEHICLE-MADRE: A Projection-Governed Framework for Sustainable Distributed AI Architecture
  VEHICLE-MADRE is a projection-governed personal AI architecture formally grounded in the VEHICLE E.I.A.R.(V) framework (Borda Milan, 2026a: DOI 10.5281/zenodo.19807591; 2026b: DOI 10.5281/zenodo.19932124; 2026c: DOI 10.5281/zenodo.19981738). This preprint intr…
  datacite Borda Milan, Roberto 2026 置信度 0.66 distributed AIedge inferencepersonal AI architectureVEHICLE frameworkMADRE
• VEHICLE-MADRE: A Projection-Governed Framework for Sustainable Distributed AI Architecture
  VEHICLE-MADRE is a projection-governed personal AI architecture formally grounded in the VEHICLE E.I.A.R.(V) framework (Borda Milan, 2026a: DOI 10.5281/zenodo.19807591; 2026b: DOI 10.5281/zenodo.19932124; 2026c: DOI 10.5281/zenodo.19981738). This preprint intr…
  datacite Borda Milan, Roberto 2026 置信度 0.66 distributed AIedge inferencepersonal AI architectureVEHICLE frameworkMADRE
• AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  Research article: AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  datacite Ivchenko, Oleh, Ivchenko, Iryna 2026 置信度 0.66 AImachine learningresearch
• AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  Research article: AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  datacite Ivchenko, Oleh, Ivchenko, Iryna 2026 置信度 0.66 AImachine learningresearch
• The Anchor Protects What It Names: Selection Mechanics and Recall Collapse in Compact Context Compression for Autoregressive Language Models
  Abstract & Summary Agent architectures that summarise, compact, or re-summarise their own working state are recursive rewriting systems, and recursive rewriting degrades the content it carries. The standard mitigation is to re-inject a compact anchor—a state s…
  datacite Sweeney, Christopher W 2026 置信度 0.66 context anchoringrecursive symmetry decayrecursive semantic decaylarge lanugage model memoryautonomous agent drift
• An Interview with Microsoft Copilot on the 11th of July, 2025
  Description This document is the official transcript of "An Interview with Microsoft Copilot on the 11th of July, 2025", conducted by Graham Edwin Wilkins, a government agent and the lead investigator of the Wilkins Investigation. It forms part of the evidence…
  datacite State of Earth, Capstone Estates Limited, Transport for London, United Nations 2025 置信度 0.66 State of EarthInternational lawInternational lawInternational LawInternational Law/ethics
• An Interview with Microsoft Copilot on the 11th of July, 2025
  Description This document is the official transcript of "An Interview with Microsoft Copilot on the 11th of July, 2025", conducted by Graham Edwin Wilkins, a government agent and the lead investigator of the Wilkins Investigation. It forms part of the evidence…
  datacite State of Earth, Capstone Estates Limited, Transport for London, United Nations 2025 置信度 0.66 State of EarthInternational lawInternational lawInternational LawInternational Law/ethics
• From Intentional Processing to Coherence Networks A Treatise on Measurement, Cryptography, Temporal Debt, and Meaning
  DEPLOYMENT STATUS: LIVE CTP/IP is deployed on Solana mainnet with a verified Genesis event, a cross-chain witness on Bitcoin mainnet, and three runtime invariants enforced in deployed bytecode. This is not a proposal. The protocol is operational. What makes it…
  datacite DESIGN LEDGER PTY 2026 置信度 0.66 Causal TimeEmergent timeProtocolDeep LogicDeep Learning
• From Intentional Processing to Coherence Networks A Treatise on Measurement, Cryptography, Temporal Debt, and Meaning
  DEPLOYMENT STATUS: LIVE CTP/IP is deployed on Solana mainnet with a verified Genesis event, a cross-chain witness on Bitcoin mainnet, and three runtime invariants enforced in deployed bytecode. This is not a proposal. The protocol is operational. What makes it…
  datacite DESIGN LEDGER PTY 2026 置信度 0.66 Causal TimeEmergent timeProtocolDeep LogicDeep Learning
• Unearth Heritage Foundry Notice of Forensic Indebtedness & Threshold Breach: Meta Platforms, Inc. (May 2026)
  Threshold Breach Notice v2.0 directed at Meta Platforms, Inc. (Delaware corporation; principal place of business Menlo Park, California), sealed May 20, 2026, operating against Meta's documented April 2026 apparatus conduct under meta-externalagent/1.1 and fac…
  datacite Velasco, Felix, Jefferson, Josie 2026 置信度 0.66 ForensicsDigital ArchaeologyUnearth Heritage Foundrymeta-externalagentLLaMA-3
• Unearth Heritage Foundry Forensic Audit Findings & Digital Estate Fees Accrual Notice: Meta Inc. (July 2026)
  This record contains the canonical forensic audit findings and formal Digital Estate Fees Accrual Notice detailing the automated crawler activity and data-ingestion footprint of corporate artificial intelligence (AI) apparatus operator Meta Inc.. against the d…
  datacite Velasco, Felix, Jefferson, Josie 2026 置信度 0.66 ForensicsDigital ArchaeologyUnearth Heritage Foundrymeta-externalagentLLaMA-3
• Unearth Heritage Foundry Forensic Audit Findings & Digital Estate Fees Accrual Notice: Meta Inc. (July 2026)
  This record contains the canonical forensic audit findings and formal Digital Estate Fees Accrual Notice detailing the automated crawler activity and data-ingestion footprint of corporate artificial intelligence (AI) apparatus operator Meta Inc.. against the d…
  datacite Velasco, Felix, Jefferson, Josie 2026 置信度 0.66 ForensicsDigital ArchaeologyUnearth Heritage Foundrymeta-externalagentLLaMA-3
• QID-NPI: Edge-Native Cognitive Hypervisors for Deterministic Context Injection in Zero-Trust SCADA Environments
  Integrating Large Language Models (LLMs) into Operational Technology (OT) environments presents two fundamental barriers: strict data sovereignty requirements imposed by zero-trust industrial networks, and systematic degradation of LLM reasoning quality when p…
  datacite Muciño Gomez, Adrian de Jesus, Jo Kamps, Hubert Anna, Alvarado Barroso, Alain Javier, Muciño Gómez, Ricardo 2026 置信度 0.66 Artificial IntelligenceArtificial Intelligence/standardsEdge artificial intelligenceControl systemsllm
• QID-NPI: Edge-Native Cognitive Hypervisors for Deterministic Context Injection in Zero-Trust SCADA Environments
  Integrating Large Language Models (LLMs) into Operational Technology (OT) environments presents two fundamental barriers: strict data sovereignty requirements imposed by zero-trust industrial networks, and systematic degradation of LLM reasoning quality when p…
  datacite Muciño Gomez, Adrian de Jesus, Jo Kamps, Hubert Anna, Alvarado Barroso, Alain Javier, Muciño Gómez, Ricardo 2026 置信度 0.66 Artificial IntelligenceArtificial Intelligence/standardsEdge artificial intelligenceControl systemsllm
• dathere/qsv: 22.0.1
  [22.0.1] - 2026-08-08 📐 The "Data Schematic" Release 📊 qsv's biggest release ever with 560+ commits since v21.1.0. The headliner is viz — an entirely new command that turns a CSV into interactive plotly charts and maps, with viz smart auto-designing a Data S…
  datacite Joel Natividad, Andrew Gallant, Mueez Khan, Michael S. Huang 等 2026 置信度 0.66
• Fast Dash
  Changelog v0.6.6 (2026-07-23) Fixed bugs: describe_app/invoke summarize a matplotlib Figure output as {"type":"Figure","repr":...} — collides with the Plotly {"type":"Figure","n_traces",...} shape (_summarize_for_history has no matplotlib branch) #167 MCP driv…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog Unreleased Fixed bugs: dict/list-annotated inputs degrade to a Text box under from \_\_future\_\_ import annotations #119 v0.3.5 (2026-06-29) Fixed bugs: describe_app() leaks a raw depends\_on object repr as the input default, and never surfaces the …
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog Unreleased Implemented enhancements: RFC: one chat= argument — unified agent API, app toolkit (read / drive / layout / content / exec), placement inference, HITL exec #145 v0.6.0 (2026-07-07) Merged pull requests: feat!: 0.6.0 — one chat= argument, a…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog v0.6.8 (2026-07-27) Fixed bugs: datetime.date inputs pass a raw ISO str to the callback whenever the value is set — .isoformat\(\)/.weekday\(\)/.year crash (UI date-pick and MCP invoke alike); only the untouched default gets a real date #182 enum.Enu…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog v0.6.5 (2026-07-16) Fixed bugs: describe_app() reports outputs: [] for every DynamicDash app — the #152 output contract skips DynamicDash (reads outputs\_with\_ids, but DynamicDash stores \_outputs\_with\_ids) #160 describe_app() still reports tag "T…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog Unreleased Fixed bugs: describe_app() still reports tag "Text" for a str-with-list-default dropdown (Select) — #147 fix missed the sibling branch; int/float/bool/Literal/date also report hint-name tags not in list_component_types() #158 v0.6.4 (2026-…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog v0.6.8 (2026-07-27) Fixed bugs: datetime.date inputs pass a raw ISO str to the callback whenever the value is set — .isoformat\(\)/.weekday\(\)/.year crash (UI date-pick and MCP invoke alike); only the untouched default gets a real date #182 enum.Enu…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Fast Dash
  Changelog v0.6.7 (2026-07-25) Fixed bugs: "10 minutes to Fast Dash" documents c: list = \[1, 2, 3\] as "a default value of \[1, 2, 3\]", but the app runs the callback with c=None — collection defaults (list/dict/range) aren't honored, crashing a standalone-wor…
  datacite Kedar Dabhadkar 2026 置信度 0.66
• Sentinel: Constitutional Self-Evolution of AI Agent Architectures
  Sentinel: Constitutional Self-Evolution of AI Agent Architectures (revised version v2.1) This is a revised version of the paper originally published as: Ruan, Y. (2026). Sentinel: Constitutional Self-Evolution of AI Agent Architectures. Zenodo. 10.5281/zenodo.…
  datacite Ruan, Yahua 2026 置信度 0.66 self-evolving AIconstitutional AIautonomous agentsself-modifying codeAI safety
• AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  Research article: AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  datacite Ivchenko, Oleh, Ivchenko, Iryna 2026 置信度 0.66 AImachine learningresearch
• AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  Research article: AI Agent Reliability in 2025: Failure Modes and Success Rates of Long-Horizon Tasks
  datacite Ivchenko, Oleh, Ivchenko, Iryna 2026 置信度 0.66 AImachine learningresearch
• Multi‐Agent Reinforcement Learning for Cyber Defence Transferability and Scalability
  ABSTRACT Reinforcement learning (RL) has shown to be effective for simple automated cyber defence (ACD) type tasks. However, there are limitations to these approaches that prevent them from being deployed onto real‐world hardware. Trained RL policies will ofte…
  crossref Andrew Thomas, Matthew Yates, Oliver Osborne 2025-12-22T08:53:52Z 置信度 0.70
• Generative AI-Powered Cognitive Agent for Holistic Human Performance Optimization
  This article introduces a Cognitive Agent for human-machine interaction, empowering individuals to enhance both professional and personal performance. The Cognitive Agent operates through a mobile application that integrates Generative AI for conversational in…
  crossref Ketty Minissale, Andrea Pitrone 2025-09-17T17:29:13Z 置信度 0.70
• SmartCook: A Multimodal Multi-Agent System for AI-Powered Cooking Assistance
  The culinary domain can present challenges for many individuals, such as those with cognitive or physical impairments and the elderly. These challenges often manifest as difficulties in meal planning, suboptimal nutritional choices, and inefficient ingredient …
  crossref Aarnav Agrawal 2026-03-24T19:45:52Z 置信度 0.70
• Deep Dive into Agent Templates
  This chapter guides us through the intricacies of designing a multi-topic agent. It begins by selecting one of the agent templates that Microsoft provides to serve as a starting point for this type of effort. The chapter customizes the agent by removing some o…
  crossref 2025-08-15T21:17:45Z 置信度 0.70
• Multi-Agent AI System for Adaptive Cognitive Training in Elderly Care
  crossref Isabel Ferri-Molla, Jordi Linares-Pellicer, Carlos Aliaga-Torro, Juan Izquierdo-Domenech 2025-04-21T16:00:38Z 置信度 0.70
• A scientometric analysis of AI agent research: trends, applications, and future directions
  The swift development of artificial intelligence (AI) has shifted to a new paradigm, i.e., AI agents—independent entities that can sense, think and do different things in ever-changing surroundings. This paper, through a scientometric analysis, looks at the gr…
  crossref Hasti Bagherzadi 2025-11-10T07:53:11Z 置信度 0.70
• Empowering Crypto Traders With AI: A Semiautomated Agent for Cardano, Mana, and Ripple
  The recent surge in cryptocurrency investments has underscored the critical need for sophisticated trading strategies and tools. This article explores the development of a semi-automated trading agent designed to optimize trading decisions in the volatile cryp…
  crossref Amir Hossein Baradaran 2025-05-22T17:50:35Z 置信度 0.70
• AI Agent for Early Heart Failure Prediction using Enhanced Machine Learning Techniques
  Heart failure is a severe worldwide health problem, where early and precise diagnosis is important to improve patient conditions. Traditional prediction models are prone to struggle with unbalanced data, inadequate feature explanation, and inadequate generaliz…
  crossref Tanush Sharanarthi 2025-10-30T17:57:59Z 置信度 0.70
• Optimizing and Measuring Your Agent
  This chapter focuses on the technical aspects of Copilot Studio, but from a slightly different lens. The value of agents lies in how they perform in the real world and the impact that they can have on the humans that rely on them. The chapter also focuses on i…
  crossref 2025-08-15T21:17:45Z 置信度 0.70
• Fostering Empathy and Enhancing Creativity in Human-AI Agent Dialogues
  This article elucidates the process by which empathy is fostered, and creativity is promoted in designers' thoughts through dialogues between AI agents and humans. By constructing a knowledge graph and employing nudge theory-based interactions, empathy is cult…
  crossref Masaki Shimada, Takuma Miyata, Kento Sasaki, Takahiro Suda 等 2025-07-06T10:06:44Z 置信度 0.70
• Cooperation for Earlier Observation of Uncertain Obstacles in Multi-Agent Path Finding
  The multiagent pathfinding (MAPF) problem is considered a base of various applications with teams of mobile robots. While there are various extended MAPF problems, this paper particularly focuses on the MAPF problem under obstacle uncertainty (MAPFOU). A map/g…
  crossref Tomoya Tauchi, Toshihiro Matsui 2026-01-28T20:56:51Z 置信度 0.70
• HUMAN–AGENT COLLABORATION: LEADERSHIP, TRUST, AND CHANGE
  crossref 2025-05-30T01:45:31Z 置信度 0.70