Research papers in ultra-low-power sensing frameworks for the Internet of Things (IoT) focus on designing and implementing sensing architectures that minimize energy consumption while maintaining accurate and reliable data collection. IoT devices, especially battery-powered sensors, are often deployed in large-scale and remote environments where frequent maintenance or battery replacement is impractical. Therefore, achieving ultra-low-power operation is critical for prolonging network lifetime and ensuring sustainable performance. Researchers have explored multiple approaches to reduce energy consumption, including duty-cycling, event-driven sensing, adaptive sampling, energy-harvesting techniques, and in-network data aggregation. Hardware-level optimizations, such as the design of energy-efficient microcontrollers, low-power transceivers, and specialized sensing modules, are complemented by software-level strategies that optimize communication protocols, data compression, and local processing to reduce unnecessary transmissions. Edge and fog computing paradigms are increasingly integrated to process data near the source, enabling selective reporting and reducing network-wide energy expenditure. Machine learning and AI-based techniques are also employed for predictive sensing and anomaly detection, allowing sensors to intelligently adjust their activity according to environmental or contextual changes. Furthermore, ultra-low-power sensing frameworks are often designed in conjunction with energy-aware routing and network management protocols to optimize the entire IoT ecosystem. Despite significant progress, challenges remain in balancing sensing accuracy, responsiveness, and energy efficiency, particularly in heterogeneous IoT deployments with diverse devices and communication standards. Overall, the literature highlights that ultra-low-power sensing frameworks are a key enabler for sustainable, long-lived, and scalable IoT networks, combining hardware-software co-design, intelligent sensing strategies, and energy-aware communication mechanisms.