conftrace_

differentiable masking

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

model compression (2) erasure method (1) knowledge critical (1) input token (1) knowledge localization (1) attribution method (1) channel pruning (1) structured pruning (1) parameter efficiency (1) pre-trained language model (1)

Papers

SepPrune: Structured Pruning for Efficient Deep Speech Separation AAAI 2026 Discovering Knowledge-Critical Subnetworks in Pretrained Language Models EMNLP 2024 How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking EMNLP 2020