PaLM: Scaling Language Modeling with Pathways

  • 类型:arxiv
  • 标识:2204.02311
  • 链接:https://arxiv.org/abs/2204.02311
  • 主题:engineering
  • 主分类:llm-infra
  • 形态:method
  • 被引:8308
  • 被引来源:Semantic Scholar
  • S2被引:8308
  • OpenAlex被引:2134
  • 影响力被引:434
  • TLDR:A 540-billion parameter, densely activated, Transformer language model, which is called PaLM achieves breakthrough performance, outperforming the finetuned state-of-the-art on a suite of multi-step reasoning tasks, and outperforming average human performance on the recently released BIG-bench benchmark.
  • OpenAlex ID:W4224308101
  • OpenAlex DOI:10.48550/arxiv.2204.02311
  • DOI:10.48550/arxiv.2204.02311
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://arxiv.org/pdf/2204.02311
  • OpenAlex更新:2026-08-26
  • 待LLM分类:否
  • 成熟度:production
  • 场景:language model、scaling、pathways
  • 标题中文:PaLM:基于 Pathways 扩展语言建模
  • TLDR中文:名为 PaLM 的 540 亿参数、密集激活的 Transformer 语言模型取得了突破性性能,在一系列多步推理任务上超越了微调后的 SOTA,并在最近发布的 BIG-bench 基准上超越了人类平均水平。
  • 来源文件
  • [OpenAlex discover]
  • [S2 enrich]