AI Paper Atlas

Research map / Benchmarks / MalNet-Tiny

MalNet-Tiny leaderboard

11 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MalNet-Tiny. Best accuracy: 94.6 (GatedGCN+, ICML 2025).

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.

MalNet-Tiny · accuracy

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

#accuracy ↑MethodPaperTopic
194.6GatedGCN+
GatedGCN
Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence
ICML 2025 · Table 3
GNN architectures
294.37GSM++ (MoT)Best of Both Worlds: Advantages of Hybrid Graph Sequence Models
ICML 2025 · GNN benchmark table
GNN architectures
394.06GSSCHow Can State Space Models Enhance Machine Learning on Graphs?
TPAMI 2026 · Table 1
GNN architectures
494.02Exphormer
GraphGPS
Exphormer: Sparse Transformers for Graphs
ICML 2023 · Table 2
GNN architectures
593.67GatedGCN+PE+VN_G
GatedGCN+PE
Understanding Virtual Nodes: Oversquashing and Node Heterogeneity
ICLR 2025 · Table: Test performance on two
GNN architectures
693.62Primphormer
GPS
Primphormer: Efficient Graph Transformers with Primal Representations
ICML 2025 · Table 2
GNN architectures
793.46Performer + WIRE
GraphGPS Performer (linear attention)
Rotary Position Encodings for Graphs
ICML 2026 · Table 3
Graph self-supervised learning
893.45DAG+GraphGPS
GraphGPS
Transformers over Directed Acyclic Graphs
NeurIPS 2023 · Table: Graph classification on
Graph self-supervised learning
993.36GPS (Transformer)Recipe for a General, Powerful, Scalable Graph Transformer
NeurIPS 2022 · text
GNN architectures
1093.3EB-GNNMessage Passing on the Edge: Towards Scalable and Expressive GNNs
ICML 2026 · Table 2
GNN architectures
1 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.