Low-light image enhancement remains a persistent challenge in computer vision, where state-of-the-art models are often hampered by hardware constraints and computational inefficiency, particularly at high resolutions. While foundational architectures like transformers and diffusion models have advanced the field, their computational complexity limits their deployment on edge devices. We introduce ExpoMamba, a novel architecture that integrates a frequency-aware state-space model within a modified U-Net. ExpoMamba is designed to address mixed-exposure challenges by decoupling the modeling of amplitude (intensity) and phase (structure) in the frequency domain. This allows for targeted enhancement, making it highly effective for real-time applications, including downstream tasks like object detection and segmentation. Our experiments on six benchmark datasets show that ExpoMamba is up to 2-3x faster than competing models and achieves a 6.8\% PSNR improvement, establishing a new state-of-the-art in efficient, high-quality low-light enhancement. Source code: https://www.github.com/eashanadhikarla/ExpoMamba.

We introduce the Aging Multiverse, a framework for generating multiple plausible facial aging trajectories from a single image, each conditioned on external factors such as environment, health, and lifestyle. Unlike prior methods that model aging as a single deterministic path, our approach creates an aging tree that visualizes diverse futures. To enable this, we propose a training-free diffusion-based method that balances identity preservation, age accuracy, and condition control. Our key contributions include attention mixing to modulate editing strength and a Simulated Aging Regularization strategy to stabilize edits. Extensive experiments and user studies demonstrate state-of-the-art performance across identity preservation, aging realism, and conditional alignment, outperforming existing editing and age-progression models, which often fail to account for one or more of the editing criteria. By transforming aging into a multi-dimensional, controllable, and interpretable process, our approach opens up new creative and practical avenues in digital storytelling, health education, and personalized visualization.

The selection event algorithm introduced an interesting method of creating offspring from parents in a multi-generational manner with a period of fitness neutrality followed by an intense selection event. This algorithm did not have a focus on maintaining diversity in the population, so it had all of the same pitfalls as other algorithms lacking such a focus. However, due to the novel multi-generational growth structure a natural family tree is created in the population, allowing for an equally natural diversity maintenance to be implemented which does not require any artificial diversity constraints to be placed on the fitness function. Instead, diversity is maintained (and encouraged) in the population through the growth dynamics of each family.

We compare common, fixed population size evolutionary strategies with a strategy incorporating a growing population size and intermittent selection events. In the latter strategy, the population size grows geometrically and selection free every generation. After a fixed number of generations, a selection event occurs which kills many of the individuals in the population and reduces the population size back to an initial value. The quality of solutions and speed of this algorithm are compared using four real-valued problem domains, to common evolutionary strategy algorithms based on (μ, λ) and (μ + λ), with promising results.