SPAP combines a mixed-integer relaxation with a penalty method and alternating minimization to prune MLP columns in LLMs, reporting lower perplexity than CFSP, FLAP, SliceGPT and FASP at matched sparsity.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SPAP: Structured Pruning via Alternating Optimization and Penalty Methods
SPAP combines a mixed-integer relaxation with a penalty method and alternating minimization to prune MLP columns in LLMs, reporting lower perplexity than CFSP, FLAP, SliceGPT and FASP at matched sparsity.