The paper introduces EMAN, an optimization-driven framework that allows a multi‑task network to grow a second independent path only when persistent optimization evidence justifies it. EMAN exposes an antisymmetric growth direction through latent relative phases and monitors multiple decision signals during training to convert local optimization evidence into a structural decision. The resulting architecture allocates shared and task‑specific capacity adaptively, and experiments on controlled rank settings, PASCAL‑Context, and NYUv2 demonstrate improved performance at a competitive computational cost.
arXiv:2602. 01619v2 Announce Type: replace-cross Abstract: Unsupervised Skill Discovery (USD) aims to autonomously learn a diverse set of skills without relying on extrinsic rewards.
arXiv:2606. 20156v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have become a powerful framework for solving PDEs by embedding physical laws into differentiable objectives.
AtomicVLA is a unified planning-and-execution framework that generates task-level plans, atomic skill abstractions, and fine-grained actions for robotic manipulation. It builds a scalable atomic skill library using a Skill‑Guided Mixture‑of‑Experts (SG‑MoE) and a flexible routing encoder that assigns new skills to dedicated experts, enabling continual learning. Experiments show that AtomicVLA outperforms baseline models on both simulated and real‑world long‑horizon tasks, achieving significant improvements in task performance and learning efficiency.
arXiv:2608. 04588v1 Announce Type: cross Abstract: Agentic systems have emerged as a promising paradigm for solving complex tasks by coordinating specialized LLM-based agents.
arXiv:2607. 01942v1 Announce Type: new Abstract: LLM-based agents have shown strong potential for solving complex multi-step tasks, yet existing performance improvements often rely on either scaling to larger backbone models or task-specific fine-tuning.
arXiv:2603. 02462v2 Announce Type: replace-cross Abstract: A key challenge in developing unified neural solvers for combinatorial optimization (CO) is the efficient generalization of models from a given set of tasks to new tasks unseen during initial training.
arXiv:2606. 02027v1 Announce Type: cross Abstract: Robot learning must produce policies that generalize to new combinations of constraints, teammates, and environments.
arXiv:2606. 29082v1 Announce Type: cross Abstract: Would experience designing faster GPU kernels also help close in on a long-standing open mathematical conjecture?
arXiv:2607. 24720v1 Announce Type: cross Abstract: Multi-turn long-horizon planning is critical for foundation model agents, yet how to fundamentally improve it remains unclear.
arXiv:2606. 18676v1 Announce Type: new Abstract: Training-free neural architecture search promises efficient discovery of high-performance networks without costly training.
HySTAR is a MAPPO-based framework that addresses structural target drift in cooperative multi‑agent reinforcement learning by anchoring an overlapping sparse hypergraph as a stable high‑order value‑decomposition scaffold. It separates adaptive representation learning from a temporally consistent decomposition basis, using a spatiotemporal encoder to capture physical and task‑dependent interactions and combining temporal and structural relevance to compute agent‑specific advantages. Experiments on SMAC, GRF, Traffic Junction, and MPE show consistent improvements over MAPPO‑style, value‑factorization, and dynamic‑grouping baselines, achieving significant gains in performance and convergence speed.
arXiv:2606. 00008v1 Announce Type: new Abstract: Multi-objective molecular optimization requires searching vast chemical spaces under conflicting objectives, where early design decisions strongly constrain downstream outcomes.