Image de couverture de livre

Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Hong, Jimyung, Kim, Jaehyung
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Training-free LLM Verification via Recycling Few-shot Examples
par: Lee, Dongseok, et autres
Publié: (2025)

Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection
par: Kang, Minjae, et autres
Publié: (2026)

Improving Chemical Understanding of LLMs via SMILES Parsing
par: Jang, Yunhui, et autres
Publié: (2025)

Entropy Meets Importance: A Unified Head Importance-Entropy Score for Stable and Efficient Transformer Pruning
par: Choi, Minsik, et autres
Publié: (2025)

Hierarchical Context Merging: Better Long Context Understanding for Pre-trained LLMs
par: Song, Woomin, et autres
Publié: (2024)

DenoiseRotator: Enhance Pruning Robustness for LLMs via Importance Concentration
par: Gu, Tianteng, et autres
Publié: (2025)

Few-shot Personalization of LLMs with Mis-aligned Responses
par: Kim, Jaehyung, et autres
Publié: (2024)

Superpose Task-specific Features for Model Merging
par: Qiu, Haiquan, et autres
Publié: (2025)

Rethinking Layer-wise Model Merging through Chain of Merges
par: Buzzega, Pietro, et autres
Publié: (2025)

REAM: Merging Improves Pruning of Experts in LLMs
par: Jha, Saurav, et autres
Publié: (2026)

MaskPrune: Mask-based LLM Pruning for Layer-wise Uniform Structures
par: Qin, Jiayu, et autres
Publié: (2025)

Learning to Correct for QA Reasoning with Black-box LLMs
par: Kim, Jaehyung, et autres
Publié: (2024)

MISA: Memory-Efficient LLMs Optimization with Module-wise Importance Sampling
par: Liu, Yuxi, et autres
Publié: (2025)

Task Vector Quantization for Memory-Efficient Model Merging
par: Kim, Youngeun, et autres
Publié: (2025)

Structural Reasoning Improves Molecular Understanding of LLM
par: Jang, Yunhui, et autres
Publié: (2024)

AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
par: Lee, Chanhyuk, et autres
Publié: (2025)

LaRA: Layer-wise Representation Analysis for Detecting Data Contamination in RL Post-Training
par: Gwak, Minju, et autres
Publié: (2026)

Optimized Feature Generation for Tabular Data via LLMs with Decision Tree Reasoning
par: Nam, Jaehyun, et autres
Publié: (2024)

Scalable Model Merging with Progressive Layer-wise Distillation
par: Xu, Jing, et autres
Publié: (2025)

Týr-the-Pruner: Structural Pruning LLMs via Global Sparsity Distribution Optimization
par: Li, Guanchen, et autres
Publié: (2025)

LayerMerge: Neural Network Depth Compression through Layer Pruning and Merging
par: Kim, Jinuk, et autres
Publié: (2024)

How Many Parameters Does Your Task Really Need? Task Specific Pruning with LLM-Sieve
par: Reda, Waleed, et autres
Publié: (2025)

Multiplex Thinking: Reasoning via Token-wise Branch-and-Merge
par: Tang, Yao, et autres
Publié: (2026)

Can LLMs Generate Diverse Molecules? Towards Alignment with Structural Diversity
par: Jang, Hyosoon, et autres
Publié: (2024)

Expert Merging: Model Merging with Unsupervised Expert Alignment and Importance-Guided Layer Chunking
par: Zhang, Dengming, et autres
Publié: (2025)

Fair Feature Importance Scores via Feature Occlusion and Permutation
par: Little, Camille, et autres
Publié: (2026)

On the Importance of Step-wise Embeddings for Heterogeneous Clinical Time-Series
par: Kuznetsova, Rita, et autres
Publié: (2023)

Spread Preference Annotation: Direct Preference Judgment for Efficient LLM Alignment
par: Kim, Dongyoung, et autres
Publié: (2024)

DISP-LLM: Dimension-Independent Structural Pruning for Large Language Models
par: Gao, Shangqian, et autres
Publié: (2024)

Measuring Sample Importance in Data Pruning for Language Models based on Information Entropy
par: Kim, Minsang, et autres
Publié: (2024)

Towards Minimizing Feature Drift in Model Merging: Layer-wise Task Vector Fusion for Adaptive Knowledge Integration
par: Sun, Wenju, et autres
Publié: (2025)

Debiasing Online Preference Learning via Preference Feature Preservation
par: Kim, Dongyoung, et autres
Publié: (2025)

From SHAP Scores to Feature Importance Scores
par: Letoffe, Olivier, et autres
Publié: (2024)

Two-Stage Grid Optimization for Group-wise Quantization of LLMs
par: Kim, Junhan, et autres
Publié: (2026)

IPPRO: Importance-based Pruning with PRojective Offset for Magnitude-indifferent Structural Pruning
par: Jung, Jaeheun, et autres
Publié: (2025)

MergeBench: A Benchmark for Merging Domain-Specialized LLMs
par: He, Yifei, et autres
Publié: (2025)

Gap-K%: Measuring Top-1 Prediction Gap for Detecting Pretraining Data
par: Kwak, Minseo, et autres
Publié: (2026)

AlphaPruning: Using Heavy-Tailed Self Regularization Theory for Improved Layer-wise Pruning of Large Language Models
par: Lu, Haiquan, et autres
Publié: (2024)

Mitigating Long-Tailed Anomaly Score Distributions with Importance-Weighted Loss
par: Lee, Jungi, et autres
Publié: (2026)

Enhancing Sufficient Dimension Reduction via Hellinger Correlation
par: Hong, Seungbeom, et autres
Publié: (2024)