This paper introduces SMGI, a structural theory of general artificial intelligence, redefining the problem of learning to focus on the controlled evolution of the learning interface rather than fixed environments. It extends statistical learning theory by treating evaluation as a dynamic, evolving component crucial for general intelligence.

This paper introduces a decision-theory framework and a novel blended AI/statistical model for generating tailored, probabilistic seasonal monsoon onset forecasts. The system significantly improves forecast skill at longer lead times, enabling heterogeneous farmers to make more informed agricultural decisions under weather uncertainty, and was operationally deployed to 38 million Indian farmers in 2025.

This paper proposes an Agentic AI-driven dual spatial-scale optimization framework for Unmanned Aerial Vehicular Networks (UAVNs) deployment. It leverages exact potential games (EPGs) at different scales and integrates a large language model (LLM) to enhance utility weight generation, leading to improved energy efficiency, latency, and throughput.

This paper introduces the Hierarchical Autonomy Evolution (HAE) framework, a novel approach to categorizing security vulnerabilities in AI agents as they evolve from cognitive entities to collective societies. It details a taxonomy of threats across three levels of autonomy, highlighting critical research gaps and guiding the development of robust, multilayered defense architectures for trustworthy AI agent systems.

This paper introduces a novel neural network framework that combines a discrete noise graph diffusion model with an autoregressive solver to enhance solutions for Vehicle Routing Problems (VRPs). By learning and integrating problem constraints through a generated constraint matrix, the approach improves robustness and achieves state-of-the-art performance on various benchmarks.