Papers with Code · CC BY-SA 4.0

STL-10

아카이브 평가 117건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
µ2Net+ (ViT-L/16)	Percentage correct 99.64	A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems 2022-09-15	코드 1
kNN-CLIP	Percentage correct 99.6	Revisiting a kNN-based Image Classification System with High-capacity Storage 2022-04-03	연결 없음
Wide-ResNet-101 (Spinal FC)	Percentage correct 98.66	SpinalNet: Deep Neural Network with Gradual Input 2020-07-07	코드 1 · 코드 2 · 코드 3
CN(d=128)	Percentage correct 98.45	Toward Understanding Supervised Representation Learning with RKHS and GAN 2021-01-01	연결 없음
CN(d=64)	Percentage correct 98.36	Toward Understanding Supervised Representation Learning with RKHS and GAN 2021-01-01	연결 없음
NSRL+CN(d=128)	Percentage correct 98.36	Toward Understanding Supervised Representation Learning with RKHS and GAN 2021-01-01	연결 없음
NSRL+CN(d=32)	Percentage correct 98.34	Toward Understanding Supervised Representation Learning with RKHS and GAN 2021-01-01	연결 없음
NSRL+CN(d=64)	Percentage correct 98.24	Toward Understanding Supervised Representation Learning with RKHS and GAN 2021-01-01	연결 없음
CN(d=32)	Percentage correct 98.17	Toward Understanding Supervised Representation Learning with RKHS and GAN 2021-01-01	연결 없음
NAT-M4	Percentage correct 97.9 FLOPS 573M PARAMS 7.5M	Neural Architecture Transfer 2020-05-12	코드 1 · 코드 2
NAT-M3	Percentage correct 97.8 FLOPS 436M PARAMS 7.5M	Neural Architecture Transfer 2020-05-12	코드 1 · 코드 2
SEER (RegNet10B)	Percentage correct 97.3 PARAMS 10000M	Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision 2022-02-16	코드 1
NAT-M2	Percentage correct 97.2 FLOPS 303M PARAMS 5.1M	Neural Architecture Transfer 2020-05-12	코드 1 · 코드 2
iGPT-L	Percentage correct 97.1	Generative Pretraining from Pixels 2020-07-17	코드 1 · 코드 2 · 코드 3 · 코드 4
NAT-M1	Percentage correct 96.7 FLOPS 240M PARAMS 4.4M	Neural Architecture Transfer 2020-05-12	코드 1 · 코드 2
EnAET	Percentage correct 95.48	EnAET: A Self-Trained framework for Semi-Supervised and Supervised Learning with Ensemble Transformations 2019-11-21	코드 1 · 코드 2
VGG-19bn	Percentage correct 95.44	SpinalNet: Deep Neural Network with Gradual Input 2020-07-07	코드 1 · 코드 2 · 코드 3
Diffusion Classifier (zero-shot)	Percentage correct 95.4	Your Diffusion Model is Secretly a Zero-Shot Classifier 2023-03-28	코드 1 · 코드 2 · 코드 3 · 코드 4
FixMatch (CTA)	Percentage correct 94.83	FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence 2020-01-21	코드 1
ReMixMatch	Percentage correct 94.77	FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence 2020-01-21	코드 1
AMDIM	Percentage correct 94.5	Learning Representations by Maximizing Mutual Information Across Views 2019-06-03	코드 1 · 코드 2 · 코드 3
MixMatch	Percentage correct 94.41	MixMatch: A Holistic Approach to Semi-Supervised Learning 2019-05-06	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7 · 코드 8 · 코드 9 · 코드 10 · 코드 11 · 코드 12 · 코드 13 · 코드 14 · 코드 15 · 코드 16 · 코드 17 · 코드 18 · 코드 19 · 코드 20 · 코드 21 · 코드 22 · 코드 23 · 코드 24 · 코드 25 · 코드 26 · 코드 27 · 코드 28 · 코드 29 · 코드 30
AMDIM-L	Percentage correct 94.2	Generative Pretraining from Pixels 2020-07-17	코드 1 · 코드 2 · 코드 3 · 코드 4
ReMixMatch (K=4)	Percentage correct 93.82	ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring 2019-11-21	코드 1 · 코드 2 · 코드 3
AMDIM	Percentage correct 93.80	A Framework For Contrastive Self-Supervised Learning And Designing A New Approach 2020-08-31	코드 1 · 코드 2
ReMixMatch (K=1)	Percentage correct 93.23	ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring 2019-11-21	코드 1 · 코드 2 · 코드 3
MP*	Percentage correct 93.19	Increasing Trustworthiness of Deep Neural Networks via Accuracy Monitoring 2020-07-03	연결 없음
UDA	Percentage correct 92.34	FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence 2020-01-21	코드 1
YADIM	Percentage correct 92.15	A Framework For Contrastive Self-Supervised Learning And Designing A New Approach 2020-08-31	코드 1 · 코드 2
FixMatch (RA)	Percentage correct 92.02	FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence 2020-01-21	코드 1