The field of robotics is moving towards more advanced and nuanced manipulation and sensing capabilities. Researchers are exploring the use of deep learning and reinforcement learning to improve the accuracy and reliability of robotic systems. One key area of focus is the development of methods for estimating the shape and pose of flexible and deformable objects, as well as techniques for gentle and damage-aware manipulation. Another important direction is the creation of datasets and calibration methods for tactile sensors, which are essential for enabling robots to interact with and understand their environment. Notable papers in this area include: Estimating Continuum Robot Shape under External Loading using Spatiotemporal Neural Networks, which presents a novel approach for accurately estimating the 3D shape of flexible continuum robots. Sim-to-Real Gentle Manipulation of Deformable and Fragile Objects with Stress-Guided Reinforcement Learning, which demonstrates a vision-based reinforcement learning approach for gentle manipulation of fragile objects. Learning to Plan & Schedule with Reinforcement-Learned Bimanual Robot Skills, which introduces a hierarchical framework for integrated skill planning and scheduling. A Humanoid Visual-Tactile-Action Dataset for Contact-Rich Manipulation, which presents a new dataset for manipulating deformable soft objects. Curvature-Aware Calibration of Tactile Sensors for Accurate Force Estimation on Non-Planar Surfaces, which develops a calibration model for flexible tactile sensors on curved surfaces.