Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Model Merging Benchmark

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-task Image Classification20-task Model Merging Benchmark (14-task + EMNIST, CIFAR10, Food101, FashionMNIST, RenderedSST2, KMNIST)
Avg Absolute Accuracy94.7
30
Model Mergingmodel merging benchmark Many-shot
Average1.015
9
Showing 2 of 2 rows