Expired80% confidenceFactExact time
CoT提示将PaLM 540B在GSM8K数学基准测试中的准确率从17.9%提升至58.1%
8
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
提示词工程完全指南：从零基础到实战应用
bilibili大模型教程6/17/2026
Related Claims
Unverified在 GSM8K 基准上，标准思维链准确率约为 56%-63%（GPT-3 规模），PAL 可提升至 72% 以上72% similarUnverified在LongMemEval基准测试中，5个chunk时BM25准确率从78.8%提升到85.2%71% similarUnverifiedNCP-ArchPreview在数学推理基准GSM8K上取得了5.99分的提升70% similarUnverifiedGPT-5.4准确率为50.5%，其配对的永远说涨基准为51.6%67% similarUnverified本地运行的Gemma 26B在453次评分预测中准确率仅为49.2%，其配对的永远说涨基准准确率为53.4%66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/44810API
curl https://kongchang.com/api/v1/knowledge/claims/44810MCP
get_claim(id=44810)