No article text was recorded for this version.
parameter importance estimation
Version 1 of 1316 papers379 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Z-Pruner: Post-Training Pruning of Large Language Models for Efficiency without Retraining
- Blog: MoE-Pruner: Pruning Mixture-of-Experts Large Language Model using the Hints from Its Router
- Blog: Task-Aware Parameter-Efficient Fine-Tuning of Large Pre-Trained Models at the Edge
- Blog: Recall Distortion in Neural Network Pruning and the Undecayed Pruning Algorithm
- Blog: Pruner-Zero: Evolving Symbolic Pruning Metric from scratch for Large Language Models
- Blog: Beyond Size: How Gradients Shape Pruning Decisions in Large Language Models
- Blog: Lookahead: a Far-Sighted Alternative of Magnitude-based Pruning
- Blog: A Simple and Effective Pruning Approach for Large Language Models
- Blog: NeuroAda: Activating Each Neuron’s Potential for Parameter-Efficient Fine-Tuning
- Blog: ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior
- Blog: Single-shot Embedding Dimension Search in Recommender System
- Blog: Improved Localized Machine Unlearning Through the Lens of Memorization
- Blog: GIFT-SW: Gaussian noise Injected Fine-Tuning of Salient Weights for LLMs
- Blog: FLRC: Fine-grained Low-Rank Compressor for Efficient LLM Inference
- Blog: EGGS-PTP: An Expander-Graph Guided Structured Post-training Pruning Method for Large Language Models
- Blog: Revisit Kernel Pruning with Lottery Regulated Grouped Convolutions
- Blog: Leave it to the Specialist: Repair Sparse LLMs with Sparse Fine-Tuning via Sparsity Evolution
- Blog: Pruning Pre-trained Language Models with Principled Importance and Self-regularization
- Blog: Adaptive Dense-to-Sparse Paradigm for Pruning Online Recommendation System with Non-Stationary Data
- Blog: AWQ: Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration
- Blog: Efficient Meta-Learning for Continual Learning with Taylor Expansion Approximation
- Blog: PRILoRA: Pruned and Rank-Increasing Low-Rank Adaptation
- Blog: NeuroAda: Activating Each Neuron's Potential for Parameter-Efficient Fine-Tuning
- Blog: FIARSE: Model-Heterogeneous Federated Learning via Importance-Aware Submodel Extraction
- Blog: Compressing Explicit Voxel Grid Representations: fast NeRFs become also small
- Blog: Less Is More? Examining Fairness in Pruned Large Language Models for Summarising Opinions
- Blog: Beyond Task Vectors: Selective Task Arithmetic Based on Importance Metrics
- Blog: REAP the Experts: Why Pruning Prevails for One-Shot MoE compression
- Blog: Importance Estimation for Neural Network Pruning
- Blog: A Simple and Effective Pruning Approach for Large Language Models
- Blog: Weight-Aware Activation Sparsity with Constrained Bayesian Optimization Scheduling for Large Language Models
- Blog: PuzzleMoE: Efficient Compression of Large Mixture-of-Experts Models via Sparse Expert Merging and Bit-packed inference
- Blog: DICE: Leveraging Sparsification for Out-of-Distribution Detection
- Blog: Beyond One-Size-Fits-All Pruning via Evolutionary Metric Search for Large Language Models
- Blog: A Provably Effective Method for Pruning Experts in Fine-tuned Sparse Mixture-of-Experts
- Blog: Mini but Mighty: Finetuning ViTs with Mini Adapters
- Blog: AWP: Activation-Aware Weight Pruning and Quantization with Projected Gradient Descent
- Blog: Advancing Weight and Channel Sparsification with Enhanced Saliency
- Blog: Activation-Informed Merging of Large Language Models
- Blog: Neural Network Panning: Screening the Optimal Sparse Network Before Training
- Blog: Unlearning Bias in Language Models by Partitioning Gradients
- Blog: PRILoRA: Pruned and Rank-Increasing Low-Rank Adaptation
- Blog: Data-Driven Channel Pruning towards Local Binary Convolution Inverse Bottleneck Network Based on Squeeze-and-Excitation Optimization Weights
- Blog: Extremely Low Footprint End-to-End ASR System for Smart Device
- Blog: Dependency-Aware Semi-Structured Sparsity of GLU Variants in Large Language Models
- Blog: Learning Best Combination for Efficient N:M Sparsity
- Blog: CHESS: Optimizing LLM Inference via Channel-Wise Thresholding and Selective Sparsification
- Blog: Activation-Informed Merging of Large Language Models
- Blog: Wanda++: Pruning Large Language Models via Regional Gradients
- Blog: CUT: Pruning Pre-trained Multi-task Models into Compact Models for Edge Devices
- Blog: WINA: Weight Informed Neuron Activation for Accelerating Large Language Model Inference
- Blog: Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradients
- Blog: Fishers for Free? Approximating the Fisher Information Matrix by Recycling the Squared Gradient Accumulator
- Blog: All-in-One: A Highly Representative DNN Pruning Framework for Edge Devices with Dynamic Power Management
- Blog: Probe Pruning: Accelerating LLMs through Dynamic Pruning via Model-Probing
- Blog: Less is KEN: a Universal and Simple Non-Parametric Pruning Algorithm for Large Language Models
- Blog: Motivating Next-Gen Accelerators with Flexible $N{:}M$ Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches
- Blog: Compressing Explicit Voxel Grid Representations: fast NeRFs become also small
- Blog: QuantumSEA: In-Time Sparse Exploration for Noise Adaptive Quantum Circuits
- Blog: Winning the Lottery with Continuous Sparsification
Showing 60 of 316 papers.
History & citation
Current · Permanent link
- Activation Informed Importance for Sparse Language Models merged into this direction (67% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_parameter_importance_estimation_724561fe,
title = {parameter importance estimation},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/parameter-importance-estimation/rdr_724561fe032c4f13b9c3430ab8a454fc},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}This exact version (version 1, frozen text):
@misc{lacuna_parameter_importance_estimation_724561fe_v1,
title = {parameter importance estimation},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/parameter-importance-estimation/rdr_724561fe032c4f13b9c3430ab8a454fc/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}