Verified90% confidenceFactTime unknown
In 2012, AlexNet won the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) by a margin of nearly 11 percentage points over the second-place entry, reducing the Top-5 error rate from 26% to 15.3%
9
Sources
90%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
Unverified2012年Alex Krizhevsky等人使用AlexNet在ImageNet竞赛中将top-5错误率从26%降低到15.3%，该突破基于两块GTX 580显卡的CUDA加速训练实现76% similarUnverifiedOn the HSS (image + video) benchmark, Glimpse scored 46.36 and GPT-6 Astra scored 53.6, a gap of about 7 points between the two.59% similarUnverifiedUnder a 96-token context, ILRN's held-out test set top-1 accuracy was 27.17%, and Transformer's was 25.83%.59% similarUnverifiedArtificial Analysis gave MuseSpark Open a score of 39, tied for 17th on the overall leaderboard, 7th among open-source models, about 7 points behind the open-source SOTA, and about 13 points ahead of Nemotron and Inkling.59% similarUnverifiedWhen DINOv3 ViT-H+/16 is pruned to 50% sparsity, removing 427 million parameters, ImageNet-1k linear probe accuracy drops by only 0.8%, without weight correction or retraining.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/7296API
curl https://kongchang.com/api/v1/knowledge/claims/7296MCP
get_claim(id=7296)