AI Paper Atlas

Research map / Benchmarks / FEMNIST

FEMNIST leaderboard

12 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on FEMNIST. Best accuracy: 92.78 (FedLC, ICML 2022).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

FEMNIST · accuracy

12 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
192.78FedLCFederated Learning with Label Distribution Skew via Logits Calibration
ICML 2022 · Table: Test accuracy on variou
Personalized and heterogeneous FL
289.03FedXDS (LRP)
ResNet8
FedXDS: Leveraging Model Attribution Methods to counteract Data Heterogeneity in Federated Learning
ICCV 2025 · Table 3
Personalized and heterogeneous FL
387.99IFCA (k=2)
2 conv + 2 FC CNN, last layer trained by IFCA
An Efficient Framework for Clustered Federated Learning
NeurIPS 2020 · Table 2
Communication-efficient FL
484.58MultiSFL
ResNet18
MultiSFL: Towards Accurate Split Federated Learning via Multi-Model Aggregation and Knowledge Replay
AAAI 2025 · Table 1
Online and replay-based
584.27FedAvP
standard CNN
FedAvP: Augment Local Data via Shared Policy in Federated Learning
NeurIPS 2024 · Table 1
Personalized and heterogeneous FL
681.44FedPracDELTA
CNN
DELTA: Diverse Client Sampling for Fasting Federated Learning
NeurIPS 2023 · Table 1
Communication-efficient FL
781.16MosTMany-Objective Multi-Solution Transport
ICLR 2025 · Table: federated learning
Combinatorial and search optimization
876.6TurboSVM-FLTurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients
AAAI 2024 · Table: Achieved test metrics
Communication-efficient FL
976.1FedGWCInteraction-Aware Gaussian Weighting for Clustered Federated Learning
ICML 2025 · Table 1
Personalized and heterogeneous FL
1073.17Recycling updates
CNN
Layer-wise Update Aggregation with Recycling for Communication-Efficient Federated Learning
NeurIPS 2025 · Table 5
Communication-efficient FL
2 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.