Since British mathematician Alan Turing proposed the Turing Test in 1950, humanity has been committed to exploring machines capable of passing this test, continuously advancing toward the goal of Artificial General Intelligence (AGI) to build machines with language intelligence comparable to humans. Language is inherently a complex human expression system governed by grammatical rules, making the development of powerful Artificial Intelligence (AI) algorithms that can understand and master language a significant challenge. Over the past two decades, language modeling approaches have been widely used for language understanding and generation, including statistical language models and neural language models.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

The Large Model Family

  • Liang Lin,
  • Yang Liu

摘要

Since British mathematician Alan Turing proposed the Turing Test in 1950, humanity has been committed to exploring machines capable of passing this test, continuously advancing toward the goal of Artificial General Intelligence (AGI) to build machines with language intelligence comparable to humans. Language is inherently a complex human expression system governed by grammatical rules, making the development of powerful Artificial Intelligence (AI) algorithms that can understand and master language a significant challenge. Over the past two decades, language modeling approaches have been widely used for language understanding and generation, including statistical language models and neural language models.