提出RollArt解耦架构，实现大规模多任务Agentic RL训练，显著提升效率与扩展性。
arXiv:2512.22560v2 Announce Type: replace-cross Abstract: Agentic Reinforcement Learning (RL) trains LLMs through multi-turn interactions with environ…
提出联合训练策略与世界观模型的新方法，让语言智能体在复杂任务中表现更佳。
arXiv:2606.02388v1 Announce Type: new Abstract: Reinforcement learning (RL) improves large language model (LLM) agents by teaching them which actions …
提出CARE-RL框架，通过能力感知机制缓解强化学习中的跨域冲突，实现多任务协同优化。
arXiv:2606.00609v1 Announce Type: new Abstract: Reinforcement learning (RL) with verifiable rewards has achieved strong progress in reasoning-oriented…
视网膜影像结合可解释多任务学习，揭示微血管信号用于2型糖尿病全身风险分层，AI医疗新突破。
arXiv:2605.24913v1 Announce Type: cross Abstract: Retinal imaging provides a non-invasive window into systemic microvascular health and has emerged as…
用Koopman算子理论推导多任务深度学习的泛化边界，为理论分析提供新视角。
arXiv:2512.19199v2 Announce Type: replace-cross Abstract: The paper establishes generalization bounds for multitask deep neural networks using operato…
目标条件智能体一次性掌握所有任务，突破多任务学习效率瓶颈。
arXiv:2605.23551v1 Announce Type: cross Abstract: A goal-conditioned reinforcement learning agent exploring an environment will see a wealth of inform…
多任务微调新方法，用连续提示优化实现参数高效的大模型适配，减少数据需求。
arXiv:2605.14055v1 Announce Type: cross Abstract: Parameter-Efficient Fine-Tuning (PEFT) is widely used for adapting Large Language Models (LLMs) for …
打破传统荧光标记依赖，多任务学习与混合架构（CNN+Transformer+LLM）让无标记单细胞成像同时完成白细胞分类和蛋白表达预测，为高通量血液分析开辟了低成本、可解释的新路径。
arXiv:2605.14717v1 Announce Type: cross Abstract: Label-free single-cell imaging offers a scalable, non-invasive alternative to fluorescence-based cyt…