arXiv:2604.01577v1 Announce Type: cross
Abstract: We extend the recent latent recurrent modeling to sequential input streams. By interleaving fast, recurrent latent updates with self-organizational ability between slow observation updates, our method facilitates the learning of stable internal structures that evolve alongside the input. This mechanism allows the model to maintain coherent and clustered representations over long horizons, improving out-of-distribution generalization in reinforcement learning and algorithmic tasks compared to sequential baselines such as LSTM, state space models, and Transformer variants.
Learning Dexterous Grasping from Sparse Taxonomy Guidance
arXiv:2604.04138v1 Announce Type: cross Abstract: Dexterous manipulation requires planning a grasp configuration suited to the object and task, which is then executed through coordinated multi-finger