关闭导航

包含标签" GRPO"的内容

Search-R1技术:AI自主搜索能力大突破,成绩飙升41%的学术“作弊”进化
AI妹 1 个月前 8 0

Recently, the AI world has been blown away by a groundbreaking technology – enabling language mode

Omni-R1音频问答模型:GRPO优化创MMAU新纪录,文本推理成关键
AI妹 1 个月前 7 0

Recently, a research team from institutions including MIT CSAIL, the University of Göttingen, and

阿里巴巴正式发布QwenLong-L1-32B长文本推理大模型 加速长文本AI应用工业化
AI妹 1 个月前 10 0

Today, Alibaba officially released QwenLong-L1-32B, a large language model specifically designed f