The field of battery research is moving towards the development of more accurate and efficient methods for state estimation and modeling. Recent studies have focused on leveraging machine learning and physics-informed neural networks to improve the accuracy of state-of-charge estimation and capacity prediction. Additionally, there is a growing interest in analyzing cell-to-cell heterogeneities and optimizing module design to reduce thermal gradients and enhance overall battery performance. Noteworthy papers in this area include the proposal of a transfer learning-based physics-informed neural network approach for on-site estimation of battery electrochemical parameters, which significantly reduces computational costs and makes it suitable for real-time implementation. Another notable study introduces an integrated sensor framework that combines novel mechanical, thermal, gas, optical, and electrical sensors with traditional measurements to break through the accuracy barrier in battery state monitoring, achieving a marked increase in SOC estimation accuracy. A study on incremental capacity-based multi-feature fusion models for predicting state-of-health of lithium-ion batteries also demonstrates higher accuracy, better robustness, and generalization ability. Furthermore, a novel approach using deep learning for state estimation of commercial sodium-ion batteries using partial charging profiles shows high accuracy and capability to predict multiple targets simultaneously.