For the last few years, chain-of-thought prompting has become the central method for reasoning in large language models. By encouraging models to “think aloud,” researchers found that step-by-step ...

The rStar2-Agent framework boosts a 14B model to outperform a 671B giant, offering a path to state-of-the-art AI without ...

The UAE has sought to position itself as a global leader in AI in a bid to diversify its economy beyond crude oil dependency.

LRM has developed a powerful CoT reasoning ability through a simple yet effective RLVR paradigm. However, the lengthy output associated with it significantly increases reasoning costs and impacts ...

LRM has developed strong CoT reasoning capabilities through a simple yet effective RLVR paradigm. However, the lengthy ...

With the launch of K2 Think, the UAE is looking to enter the global race for AI domination that has been framed as a contest ...

LRM通过简单却有效的RLVR范式，培养了强大的CoT推理能力，但伴随而来的冗长的输出内容，不仅显著增加推理开销，还会影响服务的吞吐量，这种消磨用户耐心的现象被称为“过度思考”问题。

阿联酋的高性能推理模型，刚刚开源！ 智东西9月10日消息，今日上午，阿布扎比穆罕默德·本·扎耶德人工智能大学（MBZUAI）联合AI创企G42推出的新低成本推理模型K2 Think，相关论文已arXiv预印本平台发表，昨天下午模型在Hugging ...