One Size Does Not Fit All: Distribution-Guided Tensor-Wise Adaptive Sparsification for Model Merging

arXiv:2508.06163 · cs.CL, cs.AI, cs.LG · Submitted 2025-08-08 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2025-08-08

Updated: 2026-10-13

Terminology

• Model Merging — 3× in this paper · explained in 2 episodes
• Sparsity — 2× in this paper · explained in 4 episodes
• Core — 1× in this paper · explained in 2 episodes

Sources

• Efficient Prompting Methods for Large Language Models: A Survey
• LEWIS (LayEr WIse Sparsity) -- A Training Free Guided Model Merging Approach
• DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling
• Big Transfer (BiT): General Visual Representation Learning
• AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
• GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
• A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Foundations of Large Language Models
• Scalable Model Merging with Progressive Layer-wise Distillation
• Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities
• CABS: Conflict-Aware and Balanced Sparsification for Enhancing Model Merging
• Merging Smarter, Generalizing Better: Enhancing Model Merging on OOD Data

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering