analiza aiqbench
This paper introduces a unified evaluation framework designed to assess the trustworthiness of LLMs, agentic systems, and multimodal models across eight critical dimensions including safety, robustness, and governance. By integrating meta-evaluation layers that verify the reliability of assessments themselves, the approach aims to prevent aggregate benchmark scores from obscuring critical failures in complex AI deployments. The work directly addresses the need for interpretable evidence and alignment with emerging regulatory standards like those of the EU, offering a structured path toward credible AI oversight.
A Unified Evaluation Framework for Trustworthy Large Language Models, Agentic AI, and Multimodal Systems
arXiv:2609.19524v1 Announce Type: new Abstract: Benchmark scores alone provide an incomplete basis for assessing the trustworthiness of modern artificial intelligence systems. Large language models (LLMs), agentic systems, and multimodal models (MLLMs) require different forms of assessment, yet their evaluation evidence must remain interpretable for development and oversight. We propose a unified framework that connects output-level, trajectory-level, and cross-modal assessment through eight trustworthiness dimensions: capability, robustness, safety, fairness, transparency, governance, oversight, and efficiency. The framework preserves system-specific metrics while mapping native measurements to common performance bands, accompanied by uncertainty estimates and traceable evidence. A meta-evaluation layer examines the validity, reliability, and reproducibility of the evaluation itself. Multidimensional profiles expose strengths and weaknesses, while safety-critical overrides prevent aggregate scores from masking critical failures. Mappings to governance frameworks, international standards, and European Union regulatory requirements connect technical assessment with oversight needs. The framework provides a structured basis for assessing both system performance and the credibility of the evidence supporting it, with empirical validation across deployment contexts remaining an essential next step.
Abstract:Benchmark scores alone provide an incomplete basis for assessing the trustworthiness of modern artificial intelligence systems. Large language models (LLMs), agentic systems, and multimodal models (MLLMs) require different forms of assessment, yet their evaluation evidence must remain interpretable for development and oversight. We propose a unified framework that connects output-level, trajectory-level, and cross-modal assessment through eight trustworthiness dimensions: capability, robustness, safety, fairness, transparency, governance, oversight, and efficiency. The framework preserves system-specific metrics while mapping native measurements to common performance bands, accompanied by uncertainty estimates and traceable evidence. A meta-evaluation layer examines the validity, reliability, and reproducibility of the evaluation itself. Multidimensional profiles expose strengths and weaknesses, while safety-critical overrides prevent aggregate scores from masking critical failures. Mappings to governance frameworks, international standards, and European Union regulatory requirements connect technical assessment with oversight needs. The framework provides a structured basis for assessing both system performance and the credibility of the evidence supporting it, with empirical validation across deployment contexts remaining an essential next step.
Artificial Intelligence (cs.AI)
arXiv:2609.19524 [cs.AI]
(or arXiv:2609.19524v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.19524 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 257 cuvinte
Subiecte conexe
Deconfusing Corrigibility
AI Alignment Forum · 17 oct. 2026
Trustworthy Agentic AI: Failure Modes, Mitigation Strategies, and a Lifecycle Framework for Autonomous LLM Systems
arXiv · 22 sept. 2026
2026 BAIR Graduate Showcase
BAIR Blog · 1 iul. 2026
[AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%
Latent Space · 23 sept. 2026