Lintao Ma

Long-horizon agent RL frequently stalls because fixed rewards cannot track shifting bottlenecks—co-evolving evaluation rubrics with targeted exploration skills directly breaks the repetitive failure loops of current policies.

Ling-2.6 and Ring-2.6 achieve unprecedented efficiency in agentic intelligence, enabling instant responses and deep reasoning at trillion-parameter scale.

LLMs can now forecast time series data with greater accuracy and interpretability by fusing their reasoning capabilities with traditional time series models in a novel agentic framework.