Exploring $\ell_0$ Sparsification for Inference-free Sparse Retrievers
Open paper
Benchmark
Model
Rank
Results
zero-shot-on-beir-inference-free-model-on-beir
$\ell_0$ Mask
#1
NCDG@10: 50.43
zero-shot-on-beir-inference-free-model-on-beir
$\ell_0$ Mask-$\ell_0$ Activation
#2
NCDG@10: 50.28
Rank counts only results with a code link.