Unverified50% confidenceBenchmarkExact time
EfficientNet 从 B0 到 B7，模型参数从 5.3M 增长到 66M，Top-1 精度从 77.1% 提升到 84.3%
1
Sources
50%
Confidence
Long-term
Relevance
8/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedNVFP4量化方案相比BF16基线实现了1.5倍的推理加速，同时在top-1准确率上保留了92%到97%的水平70% similarUnverifiedGLM-5.3的智能体编程准确率随投入词元增加而提升：低投入约24.7，高投入31.5，最大投入34.769% similarUnverifiedLFM2.5在模拟真实Agent场景的电信测试中提升了74个点，指令跟随IFEval达到91.8，数学推理Math500达到88.866% similarVerifiedCoT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%65% similarUnverified该模型的成功率比使用激光雷达与深度相机的多传感器系统还高 4.5 个百分点65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/810178API
curl https://kongchang.com/api/v1/knowledge/claims/810178MCP
get_claim(id=810178)