…Wu , Abstract Kairos is a world model framework that learns from diverse experiences, maintains persistent states through hybrid temporal attention mechanisms, and operates efficiently across different hardware platforms for physical AI applications…
…a Lightweight, Versatile, and Asymmetric Neural Codec Design for Real-time Operation Published on May 7 Submitted by Dan Jacobellis on May 11 Authors: Dan Jacobellis , Abstract LiVeAction is a lightweight neural…
…problems are related through the model's latent representation space , and sampling decisions can steer how learning signals evolve across that space. To operationalize this perspective, we introduce Bayesian Manifold Curriculum (BMC…
…STRIDE learns lightweight " steering operators " that mimic the behavioral shift caused by training on data subsets. By measuring how these operators perturb test predictions, we recover individual training example influences via sparse…
…Carving World Action Modeling at the Event Joints Published on Jun 1 Submitted by Ruili on Jun 3 Authors: , , , , , , , , , , , , , , , , , , , , , Abstract WALL-WM advances video-action learning by using semantic events as learning…
…enhanced by learnable expert-wise scaling , which adaptively balances the contributions of routed and shared experts. Furthermore, we introduce NormSiLU , an activation function that normalizes inputs prior to SiLU operators, producing a…
…Sparse Autoencoder Feature Distillation for Continual Learning of Large Language Models (2026) GRLO: Towards Generalizable Reinforcement Learning in Open-Ended Environments from Zero (2026) When Can LLMs Learn to Reason with Weak…
…Building on this, our central contribution, a Dual-Grained Efficiency-Aware Reinforcement Learning framework, operates at two levels. At the macro level, we propose TRACE (Tool-use Reference-Adaptive Cost Efficiency), a…
…Siyuan Guo , , , , Abstract Deployment-time learning enables large language model agents to adapt continuously during operation through episodic memory and contextual bandit optimization, improving performance across diverse tasks. AI-generated summary Large…
…Takeaway For search agents, the model is not the whole learning system. The harness — memory layout, action space, curation interface, verification records, and context rendering — is part of what RL learns to…