AI Paper Atlas

Research map / Benchmarks / psMNIST

psMNIST leaderboard

19 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on psMNIST. Best accuracy: 98.55 (LMUFormer, ICLR 2024).

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.

psMNIST · accuracy

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

#accuracy ↑MethodPaperTopic
198.55LMUFormerLMUFormer: Low Complexity Yet Powerful Spiking Model With Legendre Memory Units
ICLR 2024 · psMNIST results table
Attention and transformers
298.51LongSpikeLongSpike: Fractional Order Spiking State Space Models for Efficient Long Sequence Learning
NeurIPS 2026 · Table 6
Spiking networks
398.49Our ModelParallelizing Legendre Memory Unit Training
ICML 2021 · Table (psMNIST results)
RNNs and normalization
498.4SpikingSSM
S4D
SpikingSSMs: Learning Long Sequences with Sparse and Parallel Spiking State Space Models
AAAI 2025 · sMNIST and psMNIST results tab
Spiking networks
598.4P-SpikeSSMP-SPIKESSM: HARNESSING PROBABILISTIC SPIKING STATE SPACE MODELS FOR LONG-RANGE DEPENDENCY TASKS
ICLR 2025 · psMNIST results table
Spiking networks
698.4UnICORNN (L=3, 256 units)UnICORNN: A recurrent model for learning very long time dependencies
ICML 2021 · Table: test accuracies on perm
RNNs and normalization
798.2D-RFDendritic Resonate-and-Fire Neuron for Effective and Efficient Long Sequence Modeling
NeurIPS 2025 · Table 1
Spiking networks
897.6wRNN (16c) + MLPTraveling Waves Encode The Recent Past and Enhance Sequence Learning
ICLR 2024 · Table: sMNIST & psMNIST test a
RNNs and normalization
997.36DeepSITHDeepSITH: Efficient Learning via Decomposition of What and When Across Time Scales
NeurIPS 2021 · Table 2
Causal and generative time series
1097.3coRNN (256 units)Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependencies
ICLR 2021 · Table: Test accuracies on sMNI
RNNs and normalization
9 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.