相关论文: The Kerimov-Alekberli Model: An Information-Geomet…
As large language models (LLMs) are increasingly deployed in high-stakes and operational settings, evaluation strategies based solely on aggregate accuracy are often insucient to characterize system reliability. This study proposes a…
This paper deals with the problem of identifying and estimating dynamical parameters of continuous-time quantum open systems, in the input-output formalism. First, we characterise the space of identifiable parameters for ergodic dynamics,…
We propose that unconstrained artificial intelligence obeys a Second Law analogous to thermodynamics, where ethical entropy, defined as a measure of divergence from intended goals, increases spontaneously without continuous alignment work.…
As LLM-based systems increasingly operate as agents embedded within human social and technical systems, alignment can no longer be treated as a property of an isolated model, but must be understood in relation to the environments in which…
As autonomous systems grow more advanced, objective metrics to evaluate their ethical and legal compliance are critical for informing end users of their limitations and ensuring accountability of those who misuse them. Current ethical…
We propose an information-geometric signature of nonconservative driving that detects violations of detailed balance using the Kullback--Leibler divergence and the Fisher information. For Markov jump processes satisfying detailed balance,…
The rapid integration of AI algorithms in safety-critical applications such as autonomous driving and healthcare is raising significant concerns about the ability to meet stringent safety standards. Traditional tools for formal safety…
AI safety is still largely framed as alignment: training models to follow human preferences, safety policies, and normative constraints. That framing has improved the behavior of modern language models, but aligned behavior does not by…
Recent advancements have revealed new links between information geometry and classical stochastic thermodynamics, particularly through the Fisher information (FI) with respect to time. Recognizing the non-uniqueness of the quantum Fisher…
With the increasingly widespread adoption of AI in healthcare, maintaining the accuracy and reliability of AI models in clinical practice has become crucial. In this context, we introduce novel methods for monitoring the performance of…
Benchmarks are the primary tool for assessing progress in artificial intelligence (AI), yet current practice evaluates models on isolated test suites and provides little guidance for reasoning about generality or autonomous…
As frontier AI systems advance toward transformative capabilities, we need a parallel transformation in how we measure and evaluate these systems to ensure safety and inform governance. While benchmarks have been the primary method for…
Security evaluations inherently depend on stable identifiers. Any finding, audit, or regulatory decision must remain attached to the specific artifact it pertains to. Continuously updated artificial intelligence systems violate this core…
When AI interacts with the physical world -- as a robot or an assistive agent -- new safety challenges emerge beyond those of purely ``digital AI". In such interactions, the potential for physical harm is direct and immediate. How well do…
Recent AI systems compress the distance between capability growth and capability deployment. Earlier high-risk technologies were slowed by capital intensity, physical bottlenecks, organizational inertia, and specialized supply chains. By…
The training algorithms for AI systems all introduce far-from-equilibrium dynamical processes, and understanding the irreversibility of these algorithms is a fundamental step towards understanding the learning dynamics of modern AI systems.…
This paper develops KL-Ergodic Exploration from Equilibrium ($\text{KL-E}^3$), a method for robotic systems to integrate stability into actively generating informative measurements through ergodic exploration. Ergodic exploration enables…
The consistency of AI-native applications depends on the behavioral consistency of the model endpoints that power them. Traditional reliability metrics such as uptime, latency and throughput do not capture behavioral change, and an endpoint…
As AI models scale to billions of parameters and operate with increasing autonomy, ensuring their safe, reliable operation demands engineering-grade security and assurance frameworks. This paper presents an enterprise-level, risk-aware,…
Optimum designs for parameter estimation in generalized regression models are standardly based on the Fisher information matrix (cf. Atkinson et al (2014) for a recent exposition). The corresponding optimality criteria are related to the…