Pruning the most attribution-prominent MLP neurons in a single layer, chosen via a 10-sample validation sweep, consistently improves multiple-choice accuracy across four instruction-tuned LLMs.
will you find these shortcuts? a protocol for evaluating the faithfulness of input salience methods for text classification
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Detecting and Pruning Prominent but Detrimental Neurons in Large Language Models
Pruning the most attribution-prominent MLP neurons in a single layer, chosen via a 10-sample validation sweep, consistently improves multiple-choice accuracy across four instruction-tuned LLMs.