Pointer Sentinel Mixture Models

  • 类型:arxiv
  • 标识:1609.07843
  • 链接:https://arxiv.org/abs/1609.07843
  • 主题:rag
  • 主分类:llm-infra
  • 形态:method
  • 被引:4466
  • 被引来源:Semantic Scholar
  • S2被引:4466
  • OpenAlex被引:484
  • 影响力被引:685
  • TLDR:The pointer sentinel-LSTM model achieves state of the art language modeling performance on the Penn Treebank while using far fewer parameters than a standard softmax LSTM and the freely available WikiText corpus is introduced.
  • OpenAlex ID:W2525332836
  • OpenAlex DOI:10.48550/arxiv.1609.07843
  • DOI:10.48550/arxiv.1609.07843
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://arxiv.org/pdf/1609.07843
  • OpenAlex更新:2026-08-06
  • 待LLM分类:否
  • 成熟度:research
  • 场景:语言建模、指针网络
  • 标题中文:Pointer Sentinel Mixture Models
  • TLDR中文:Pointer sentinel-LSTM 模型在 Penn Treebank 上以远少于标准 softmax LSTM 的参数量达到 SOTA 语言建模性能,并开源了 WikiText 语料库
  • 来源文件
  • [OpenAlex discover]
  • [S2 enrich]