AI governance Trust Under Scrutiny: Mira Murati’s Testimony and the Future of AI Governance Explore Mira Murati's sworn testimony alleging Sam Altman's dishonesty regarding AI safety, and its implications for leadership, ethics, and trust in the rapidly evolving AI industry.
Embodied AI safety Ensuring Safety in Embodied AI: A Comprehensive Look at Risks, Attacks, and Defenses Explore the critical safety challenges of Embodied AI operating in physical environments, from adversarial attacks to human-robot interaction risks, and discover robust defense strategies.
Aviation Safety Building Trust in the Skies: How AI and Knowledge Graphs Revolutionize Aviation Safety Explore a novel framework combining Large Language Models with Knowledge Graphs to create verifiable, trustworthy AI for aviation safety, mitigating hallucination and ensuring regulatory compliance.
Explainable AI Explainable AI for Human Activity Recognition: Building Trust in Intelligent Systems Explore Explainable AI (XAI) for Human Activity Recognition (HAR). Understand how transparent AI models enhance trust, improve reliability, and unlock new applications in healthcare, smart cities, and industry.
Industrial Time Series Forecasting Dual-Stream Physics-Residual Networks: Enabling Trustworthy AI for Industrial Forecasting Explore DSPR, a novel AI framework that combines data-driven accuracy with physical plausibility for industrial time series forecasting. Learn how it enhances reliability and interpretability in critical operations.
Membership Inference Attack Safeguarding Enterprise AI: How Advanced Attacks Like ReproMIA Drive Proactive Privacy Auditing Explore ReproMIA, a novel framework leveraging model reprogramming to proactively detect privacy vulnerabilities in AI models. Learn how it enhances data security for LLMs, Diffusion Models, and more.
Bimodal Regression Enhancing Trustworthy AI: Distribution-Aware Loss Functions for Robust Bimodal Regression Discover how new distribution-aware loss functions improve AI models' predictive accuracy and trustworthiness by robustly handling bimodal data, outperforming traditional methods.
LLM uncertainty estimation Unveiling LLM Uncertainty: How Cross-Layer Insights Combat AI Hallucinations Explore a novel method for Large Language Models to estimate uncertainty using intra-layer information, improving reliability, transferability, and robustness in critical AI applications.
Multi-evidence reasoning Advancing Trustworthy AI: Formal Guarantees for Multi-Evidence Reasoning with Latent Posterior Factors Explore Latent Posterior Factors (LPF), a principled AI framework with formal guarantees for combining diverse, noisy evidence in high-stakes applications like healthcare and finance, ensuring accuracy, robustness, and interpretability.
