The field of clustering and graph embeddings is witnessing significant developments, with a focus on improving the accuracy and efficiency of algorithms. Researchers are exploring new approaches to clustering, such as learning-augmented streaming algorithms and chromatic correlation clustering, which have shown promising results in terms of approximation ratios and space efficiency. Additionally, there is a growing interest in incorporating negative statements into knowledge graph embeddings, which has been shown to improve predictive performance. The development of absolute indices for determining compactness, separability, and the number of clusters is also a notable trend, as it provides a more robust and reliable way to evaluate clustering solutions. Noteworthy papers include:

• One that presents a novel approach to learning-augmented streaming algorithms for correlation clustering, achieving better-than-3 approximation under good prediction quality.
• Another that introduces a dual-model architecture for knowledge graph embeddings, which improves predictive performance by integrating explicitly declared negative statements.