quality gate definition

This work addresses the challenge of reconciling task-level verification and regulatory traceability within high-velocity AI-assisted engineering workflows. The authors propose an “infinite loop” framework that integrates agile iteration with V-model validation, embedding independent verification and compliance auditing into every development cycle through a multi-agent AI architecture. The system automatically generates audit-ready documentation and incorporates critical human-in-the-loop approval gates. By natively embedding compliance capabilities into the development process, the approach achieves 100% requirement-level verification and enables trustworthy delivery with minimal human intervention. In a hardware-in-the-loop case study, the system attained full requirement pass rates with an average of only six human prompts per cycle, demonstrating a projected cost reduction of 10–50× compared to conventional methods.

This study addresses the pervasive lack of structural integrity in current AI governance documents, which often fail to meet critical requirements such as traceability, dynamic re-verification, and objective evidence. To bridge this gap, the work systematically adapts structural governance principles from aviation software certification standards (DO-178C/DO-330) and proposes a novel integrity framework tailored for static AI governance artifacts. The framework introduces three key concepts—“epoch constraints,” “proof surfaces,” and “structural gaps”—and establishes the seven-principle PromptQ system. Structural analysis of mainstream governance documents reveals that 37% fall below a basic quality threshold, thereby demonstrating the framework’s effectiveness and practicality in enhancing the rigor and verifiability of AI governance documentation.

本文提出Agile-V Assurance Spine，通过权威源配置文件、工件绑定、风险适当独立性和时效性等方法解决工程生命周期中对代理输出的正当行动问题。

This study addresses the accountability deficit in agent development arising from the misalignment between platform controls and service provider terms. By analyzing four categories of tools and policy documents, we map workflow responsibilities and propose a novel grid model distinguishing verification mandates from executors. This framework reveals structural deficiencies in approval mechanisms, demonstrating that responsibility gaps have evolved from human oversight to inherent product attributes. Empirical findings indicate conflicting accountabilities across layers, contradictory attribution logic, and insufficient efficacy of approval artifacts. To support further research, we release a comprehensive dataset and validation scripts as open-source resources. Collectively, this work provides both theoretical grounding and empirical evidence necessary for reconstructing accountability frameworks in agent-based software systems, highlighting the urgent need to address systemic rather than incidental failures in current governance architectures.

This work addresses the limitations of existing AI trustworthiness assessment approaches, which are either too abstract to support full lifecycle monitoring or rely on single metrics insufficient for governance needs. The paper proposes a lightweight, auditable framework for dynamic trustworthiness management that integrates formal modeling with governance processes. By employing context-sensitive trustworthiness dimension protocols and interpretable rule learning based on decision trees, the framework enables end-to-end monitoring and documentation of AI systems—from design and deployment through re-evaluation. Novel diagnostic tools, including hierarchical transitions, margin-of-boundary analysis, and profile drift detection, are introduced alongside clearly accountable human-in-the-loop checkpoints. Experiments on synthetic AI lifecycle trajectories demonstrate the framework’s effectiveness in detecting performance degradation, abrupt perturbations, and impacts of system updates, thereby establishing a transparent, traceable, and contestable evidentiary basis for AI governance.

This study addresses the inefficiency of manual auditing in corporate governance and the challenges in validating automated alternatives by proposing a task replacement system within a Digital Governance Framework (DGF). Methodologically, it introduces a residual workload threshold as the criterion for task substitution, establishing an information sufficiency gating mechanism. Architecturally, the framework integrates intelligent agents, rule engines, and evidence services, automating auditing workflows through forward deployment engineering and constructing the DGF-Bench benchmark for multi-model evaluation. Experimental results demonstrate that models such as Gemini achieve success rates up to 94.98% under strict gating conditions, confirming the technical feasibility of automating specific governance tasks and offering an effective paradigm for enterprise digital governance.

This work addresses the inadequacy of existing large language model (LLM) lifecycle frameworks, which predominantly emphasize operational efficiency while lacking explicit support for security-critical activities—such as data provenance, component signing, and access control—and failing to align governance requirements with specific lifecycle phases. The paper proposes the first security-oriented LLM system lifecycle model, structured not by workflow but by security boundaries, organizing 32 phases into four layered pipelines: data, model, distribution, and application, while integrating LLMOps and governance pillars. It uniquely identifies 13 distinct security-critical phases and exposes a structural imbalance wherein regulatory evidence is concentrated at deployment despite pivotal decisions occurring during development. By mapping key standards—including NIST AI RMF, the EU AI Act, and ISO/IEC 42001—the study establishes a phase-to-governance correspondence mechanism, yielding a comprehensive, lifecycle-spanning security analysis framework that offers structured guidance for compliance and secure design.

Current AI-assisted scientific writing lacks auditable generation processes and mechanisms for accountability, undermining the verifiability of research credibility and compliance. This work proposes a novel auditing paradigm embedded directly within the production workflow, enforcing end-to-end traceability, immutability, and third-party reproducibility of AI involvement through preregistered blind-spot indicator cards, sealed execution environments, and automated gatekeeping intercepts. Core technical components include Git-sealed lineage anchoring, hash-bound provenance tracking, red-flag interception protocols, cross-model role isolation, and programmatic assembly. In experimental validation, one project was automatically terminated when preregistered confirmatory tests triggered a No-Go decision. An open-source toolkit is released to enable independent recomputation of all core audit metrics by third parties.

研究探讨了过程追踪对大型语言模型监督者决策标准的影响，使用信号检测理论分析五种模型在19个任务中的表现，发现详细的过程追踪会增加错误警报。

This study addresses the breakdown of end-to-end guarantees caused by protocol-agnostic design in decentralized agent economies. We introduce "guarantee closure" as a novel task-relative criterion that decouples the evaluation of receipt soundness and completeness. By constructing a six-phase, seventeen-category attribute taxonomy spanning the full task lifecycle, and integrating formal analysis, controlled workflow execution, and large-scale exhaustive verification, this work systematically repairs failure points between verification and settlement. Through 840 matched executions and over ten thousand test cases, we precisely identify and quantify specific failure modes. Ultimately, this research provides both a theoretical foundation and a practical framework for coordinating mechanisms across heterogeneous systems.