AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Automatic Pruning Discovery for Large Language Models

Haidong Kang, Lihong Lin, Enneng Yang, Hong-Ning Dai, Hao Wang

ICML 2026 · Large language models › Language, safety and interpretability · approach: LLM unlearning and benchmarks · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add step-by-step CoT to LLM pruning-rule design (ablation)AutoPrune search · LLaMA-2 7B WikiText-2perplexity: 16.55 -> 7.17Improved
add graph-driven multi-path CoT over linear CoT (ablation)AutoPrune search · LLaMA-2 7B WikiText-2perplexity: 7.17 -> 6.28Improved
add SDSA to Wanda (ablation)Wanda · LLaMA-1 7B WikiText-2, 60% sparsityperplexity: 10.66 -> 9.88Improved
add SDSA to SparseGPT (ablation)SparseGPT · LLaMA-1 7B WikiText-2, 60% sparsityperplexity: 10.51 -> 8.85Improved
add SDSA to AutoPrune (ablation)AutoPrune · LLaMA-2 7B WikiText-2, 60% sparsityperplexity: 8.93 -> 8.88No clear effect
replace uniform sparsity with SDSA (ablation)Wanda · LLaMA-1 7B WikiText-2, 70% sparsityperplexity: 85.77 -> 66.93Improved
replace OWL allocation with SDSA (ablation)Wanda · LLaMA-1 7B WikiText-2, 70% sparsityperplexity: 75.06 -> 66.93Improved
use global sparsity allocation instead of uniform (ablation)Wanda · LLaMA-1 7B WikiText-2perplexity: 7.26 -> 14848 at 50%Hurt
vary number of calibration samples 16 to 2048 (ablation)AutoPrune · LLaMA-1 7B WikiText-2, 50%perplexity: 7.12 (16) vs 7.11 (2048)No clear effect
use AutoPrune-discovered rule instead of WandaWanda · LLaMA-1 7B WikiText-2, 60%perplexity: 10.66 -> 9.63Improved
use AutoPrune rule instead of WandaWanda · LLaMA-2 7B zero-shot, RTEaccuracy: 53.43 -> 58.12Improved
use AutoPrune rule instead of WandaWanda · LLaMA-2 13B zero-shot, RTEaccuracy: 64.02 -> 61.01Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy