Importance

Follow this topic Ask about this topic

Momentum

9 papers in the last four weeks, up 13% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 92

All topics

Don't Waste the Noise: Importance-Guided Perturbation Allocation under Joint Global and Local Constraints
Oct 1, 2026Melika Shirian, Kianoosh VadaeiAdversarial Training Quadratic Slack Penalty
Independent Researcher
Importance-Aware Feature Sparsification for Wireless Split Learning
Sep 30, 2026Bumjun Kim, Yoon Huh, Wan ChoiSparsity Balanced Learning
Department of Electrical and Computer Engineering, Seoul National University (SNU), and the Institute of New Media and Communications, SNU, Seoul 08826, Korea
Automatic Rank Allocation for Low-Rank Adaptation in Large Language Models via lp Regularization
Sep 24, 2026Zebang Xie, Chuanyang Zheng, Yik-Chung Wu +1Regularization Importance
National University of Singapore, Singapore · Huawei, Hong Kong · The University of Hong Kong, Hong Kong +1
KV-COBRA: KV Cache Compression via Co-Optimized Bit-Rank Allocation
Sep 21, 2026Sihyeon Ha, Jaeho Lee, Yo-Seb JeonKey-Value Cache Compression Cache
Pohang University of Science and Technology (POSTECH)
$λ$ -Controlled GRPO: Turning Flow-Matching Ratio Instability into a Budgeted Resource
Sep 18, 2026Yufeng Wang, Parivesh Priye, Meeshawn Marathe +1Reward Gradients Flow-Grpo
Stony Brook University · Georgia Institute of Technology · Rivian and Volkswagen Group Technologies
Null importance: Disentangling relevance for interpretable machine learning
Sep 16, 2026Garvesh Raskutti, Kris Sankaran, Jiaxin YeFeature Importance Importance
Department of Statistics, University of Wisconsin–Madison
Contiguity, Not Importance: Budgeted Repair of Stale KV Caches After Document Edits
Sep 16, 2026Mingyang Mao, Wyatt Mackey, Xiaomin LinKey-Value Cache Multimodal Knowledge Editing
University of South Florida, Electrical and Computer Engineering · DEVCOM Army Research Laboratory
Robust Policy Optimization via Adversarial Importance Sampling
Sep 14, 2026Amine Andam, Jamal Bentahar, Mustapha HedabouOffline Reinforcement Learning Adversarial Training
Mohammed VI Polytechnic University · Khalifa University · Concordia University
Pixel Decodability Is Not a Compression Signal: Causally Evaluating Importance Proxies for Visual KV-Cache Eviction
Sep 14, 2026Chenyu Zhou, Qiliang Jiang, Shuning Wu +1Data Compression Methods Pixels
School of Engineering, Institute of Science Tokyo, Japan · College of Control Science and Engineering, Zhejiang University, China · Department of Electrical and Computer Engineering, National University of Singapore
Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning
Sep 3, 2026Kevin Du, Alexander Hoyle, Laura Ruis +1Reasoning Chain Reasoning Traces
ETH Zürich · Work done during an internship at Cohere. · MIT +1
Differentiable Interval Bottlenecks for Interpretable Anomaly Detection in Numerical Data
Sep 3, 2026Lamine Diop, Marc PlantevitGraph Anomaly Detection Interpretable Anomaly Detection
EPITA, Laboratoire de Recherche de l’EPITA (LRE) · Le Kremlin-Bicˆetre, France
Residual Sparsification via Output Importance for Compressing Mixture-of-Experts LLMs
Sep 1, 2026Seungwoo Jung, Dohyeok Kwon, Seungmin Cha +4Large Language Model Compression Importance
Department of Computer Science and Engineering, Korea University · Department of Computer Science and Artificial Intelligence, Dongguk University
Conditional Visual Evidence Utility: State-Dependent Rank Reversals in Frozen Vision-Language Encoders
Aug 28, 2026Yunxuan Fang, Xinhe WangVisual Evidence Multiple Vision Tasks
Beihang University
StepKV: Step-Aware KV Cache Compression for LLM Agents
Aug 26, 2026Boyu Feng, Jiahong Liu, Yifan Li +7Key-Value Cache Compression Cache
The Chinese University of Hong Kong · Huawei Technologies Co., Ltd
On the global feature importance for interpretable and trustworthy heat demand forecasting
Aug 13, 2026Milan ZdravkovićFeature Importance Explainable Artificial Intelligence
Faculty of Mechanical Engineering University of Niš ul. Aleksandra Medvedeva 14, 18000 Niš, Serbia
Unlocking the Power of Medical Tabular Data via Semantic-Aware Multimodal Pre-training
Aug 11, 2026Yingsheng Liu, Haiming Li, Jingmin Zhu +6Multimodal Clinical Data Tabular Learning
Faculty of Information Technology, Monash University, Melbourne, Australia · Monash University, Melbourne, Victoria, Australia · The University of Queensland, Brisbane, Queensland, Australia
On the Importance of Geometric Nonlinearity and Temperature-Dependent Properties in Multi-Material Thermo-Mechanical Topology Optimization
Aug 11, 2026Shirin Hosseinmardi, Xiangyu Sun, Ramin BostanabadTopology Optimization Nonlinearity
Department of Mechanical and Aerospace Engineering, University of California, Irvine · Department of Civil and Environmental Engineering, University of California, Irvine
Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models
Aug 7, 2026Minseok Kang, Hyunwoo Kim, Chanyoung Kim +3Visual Token Pruning Large Language Model Compression
Department of Artificial Intelligence, Chung-Ang University, Seoul, Korea · School of Computer Science and Engineering, Soongsil University, Seoul, Korea · Department of Computer Science, Kookmin University, Seoul, Korea
Bayesian Data Reweighting Improves Multimodal Retrieval for Knowledge-Based Visual Question Answering
Aug 3, 2026Jingchen Sun, Shaobo Han, Ruiyi Zhang +5Knowledge-Based Visual Question Answering Multimodal Retrieval
University at Buffalo · NEC Laboratories America · Adobe Research +2
Messages, Not Tokens: Grounded Coresets for Faithful VLM Compression
Aug 3, 2026Long Qian, Jiaqi Wei, Bingke Zhu +2Large Language Model Compression Learned Image Compression
Foundation Model Research Center, Institute of Automation, Chinese Academy of Sciences, Beijing, China · School of Engineering, Cardiff University, Cardiff, United Kingdom · School of Future Technology, University of Chinese Academy of Sciences, Beijing, China +1
Model-Agnostic FDR Control via Group Gaussian Mirror and Permutation SHAP
Aug 2, 2026Jiaan Han, Junxiao Chen, Yanzhe FuFeature Importance False Discovery Rate
Columbia Business School Columbia University · Department of Statistics University of Hong Kong
Relative Parameter Importance in Task-Agnostic Replay-Free Continual Learning
Aug 1, 2026Malavika Suresh, Ikechukwu Nkisi-Orji, Nirmalie WiratungaReplay-Based Continual Learning Continual Learning
Robert Gordon University
LAST: The Last Query Token Guides Visual Token Pruning for Edge-Cloud Collaborative MLLM Inference
Jul 30, 2026Feng Yang, Xinrui Ju, Keyang Zhang +6Visual Token Pruning LLM Inference Optimization
1City University of Hong Kong · 2Peng Cheng Laboratory
Contrastive Concept Importance: Explaining Pairwise Class Decisions Through Automatically Extracted Concept Representations
Jul 30, 2026Roel Visser, Isaac Roberts, Barbara HammerContrastive Learning Explainability
Bielefeld University, Inspiration 1, 33619 Bielefeld, Germany
Structured Redundancy Modeling for Efficient Visual Token Pruning in High-Resolution MLLMs
Jul 25, 2026Jouwon Song, Woohyeong Kim, Kyeongbo KongVisual Token Pruning Multimodal Large Language Models
LG Electronics, Seoul, Republic of Korea · Pusan National University, Busan, Republic of Korea
Variable Importance Identification Through Lazy Training for Binary Classification
Jul 25, 2026Anand Singh, Luke Pennella, Eshan Kabir +1Feature Importance Explainability
Department of Mathematics, Georgia Institute of Technology · Department of Mathematics, University of Connecticut · Department of Mathematics, Columbia University +1
HiKV: Hierarchical Importance-Aware KV Cache with Hardware Acceleration for LLM Decoding
Jul 24, 2026Chao Fang, Jun Yin, Man Shi +1Key-Value Cache Large Language Model Decoding
ESAT-MICAS, KU Leuven, Leuven, Belgium · UC Berkley, CA, USA
Deconstructing Off-Policy Ratios: Entropy-Scaled Trust Regions for Asynchronous Reinforcement Learning
Jul 24, 2026Guanqun Zhao, Zijun Xie, Binbin Zheng +5Trust Region Training-Inference Mismatch
Beijing University of Posts and Telecommunications · Baidu Inc. · Peking University +2
Representing Entity Importance in AI Knowledge Systems: A Dual-Signal Framework of Audience Evaluation and Structural Authority
Jul 23, 2026Shen XuImportance Audience
Independent Researcher
Importance-Aware OBS Pruning for Diffusion Models
Jul 22, 2026Ba-Thinh Lam, Srijan Das, Hieu LeDiffusion Models Generative Image Compression
Department of Computer Science UNC Charlotte NC, USA
CausalGate: Causal Importance Distillation for Transformer Module Pruning
Jul 21, 2026Kiran Nair, Smriti Regmi, Rodrigue RizkLLM Inference Optimization Transformer Architectures
USD Artificial Intelligence Research Lab, Department of Computer Science University of South Dakota, Vermillion, SD 57069, USA
Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information
Jul 21, 2026Priyank Agrawal, Ankur Samanta, Shervin Ghasemlou +4Reinforcement Learning With Verifiable Reward Verifiable Rewards
1Meta AI · 2Columbia University · *Work done at Meta
Importance Sampling and PCA for Finding Failures in Commercial Autonomous Vehicles
Jul 20, 2026Hailey Warner, Duncan Eddy, Shreya Parjan +6Autonomous Navigation Collision Prediction
Department of Aeronautics and Astronautics, Stanford University, Palo Alto, CA · Torc Robotics, Blacksburg, VA
Kernel weighted importance sampling for off-policy evaluation in contextual bandits
Jul 16, 2026Joshua Spear, Matthieu Komorowski, Rebecca Pope +2Contextual Bandit Framework Importance Weighting
Institute of Child Health, University College London · Strive Health Ltd. · National Institute for Health Research +3
NodeImport: Imbalanced Node Classification with Node Importance Assessment
Jul 15, 2026Nan Chen, Zemin Liu, Bryan Hooi +3Imbalanced Classification Graph Neural Networks
Johns Hopkins University Baltimore, Maryland, USA · Zhejiang University Hangzhou, Zhejiang, China · National University of Singapore Singapore, Singapore +1
Attention-Free and Lightweight Token Reduction for Efficient Vision-Language Models
Jul 15, 2026Xuanyi Hao, Zuoyuan Zhang, Zhibo Wang +4Long Visual-Token Sequences Recent Vision-Language Models
State Key Laboratory of Blockchain and Data Security, Zhejiang University, Hangzhou 310027, China · Hong Kong University of Science and Technology, Hong Kong SAR, China · School of Artificial Intelligence, Nanchang University, Nanchang 330038, China
Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning
Jul 13, 2026Hao Kong, Di Liu, Xiangzhong Luo +5Convolutional Neural Networks Neural Network Inference
School of Computer Science and Engineering, Nanyang Technological University, Singapore · HP-NTU Digital Manufacturing Corporate Lab, Nanyang Technological University, Singapore · Department of Computer Science, Norwegian University of Science and Technology, Trondheim, Norway. +1
Spectral Heat Flow for Conservative Token Condensation in Vision-Language Models
Jul 12, 2026Zhaoyang Li, Yanjun Li, Wangkai Li +2Visual Token Pruning Importance
School of Information Science and Technology / National Key Laboratory of Deep Space Exploration, University of Science and Technology of China.
The Importance of Encoder Choice:A Tabular-Image Study
Jul 8, 2026Ilia Koloiarov, Diego Coello de Portugal Mecke, Vijaya Krishna Yalavarthi +2Multimodal Learning Encoders
ISMLL & VWFS DARC · University of Hildesheim · Hildesheim, Germany +1
Explainable AI for Cancer Drug Response Prediction: Beyond Univariate Feature Attributions
Jul 1, 2026Martino Ciaperoni, Margherita Lalli, Simone Piaggesi +6Explainable Artificial Intelligence Importance
KDD Lab, Scuola Normale Superiore Pisa, Italy · KDD Lab, University of Pisa Pisa, Italy · EMBL-EBI Hinxton, United Kingdom +1
DFMU: Data-Frugal Machine Unlearning
Jun 24, 2026Sajith U, Prateek KeserwaniMachine Unlearning Unlearning Method
Samsung R&D Institute India-Bangalore, India
Importance of Intent-Sharing for V2X-based Maneuver Coordination
Jun 23, 2026Rafael Molina-Masegosa, Sergei S. Avedisov, Miguel Sepulcre +3V2X Automated Vehicles
Universidad Miguel Hernandez de Elche, Elche, Spain · Toyota Motor North America R&D – InfoTech Labs, Mountain View, CA, USA
ARIA: Adaptive Region-Based Importance Allocation for Conditional Diffusion Distillation
Jun 22, 2026Loay Mualem, Vinh Tong, Samir Darouich +1Text-Conditioned Diffusion Model Knowledge Distillation
Institute for AI, University of Stuttgart · IMPRS-IS · Institute of Theoretical Chemistry, University of Stuttgart.
Predicate Importance Estimation and Decoupled Rationale-Score Distillation for Entity Alignment
Jun 22, 2026Keunha Kim, Yoonjin Jang, Hyeon-gu Lee +2Knowledge Graphs Importance
SungKyunKwan University · NAVER
Complementary Attention Head Pruning for Efficient Transformers
Jun 17, 2026Yaniv Livertovsky, Shahar Somin, Gonen SingerTransformer Attention Unstructured Pruning
Bar-Ilan University Ramat Gan, Israel
Attribution-Guided and Coverage-Maximized Pruning for Structural MoE Compression
Jun 16, 2026Yifu Ding, Jiacheng Wang, Ge Yang +4Mixture-Of-Expert Inference Mixture-Of-Experts
State Key Laboratory of Complex & Critical Software Environment, Beihang University · School of Computer Science and Engineering, Beihang University · Nanyang Technological University +1
WaveSync: Constrained Wavefront Optimization for Synchronized Co-Speech Gestures in Humanoid Robots
Jun 15, 2026Thang Tran Viet, Thanh Nguyen Canh, Gia Huy Uong +4Co-Speech Gesture Generation Importance
University of Engineering and Technology, Vietnam National University, 10000, Hanoi, Vietnam. · School of Information Science, Japan Advanced Institute of Science and Technology, Nomi, 923-1211, Ishikawa, Japan. · School of Electronics and Computer Science, University of Southampton, SO17 1BJ Southampton, United Kingdom. +1
Stepwise Token Selection for Efficient Multimodal Large Language Models
Jun 14, 2026Landi He, Shawn Young, Lijian XuVisual Token Pruning Multimodal Large Language Models
Shenzhen University of Advanced Technology
Beyond Importance: Interchange-Sobol Sensitivity Reveals Task-Specific Content Channels in Transformer Components
Jun 12, 2026Yifeng Guo, Jin-Hong Du, Xiang ChenTransformer Architectures Importance
St. Jude Children’s Research Hospital Memphis, TN · The University of Hong Kong Hong Kong, CN
Correcting Variable Importance Scored by Random Forests
Jun 9, 2026Guancheng Zhou, Haiping Xu, Jason Liu +1Feature Importance Random Forest
Mathematics and Data Science University of Massachusetts, Dartmouth, MA · Computer and Information Science University of Massachusetts, Dartmouth, MA · The Rivers School, Weston, MA
Importance-Aware Scheduling for High-Dimensional Hyperparameter Optimization
Jun 8, 2026Ruinan Wang, Ian Nabney, Mohammad GolbabaeeGeneral Grid Search Framework High-Dimensional
University of Bristol, Bristol, United Kingdom
Neuro-Symbolic Learning for Long-Horizon Task Planning Under Complex Logical Constraints
Jun 5, 2026Qiwei Du, Zitong Zhan, Shaoshu Su +7Classical Planning Neuro-Symbolic Framework
Spatial AI & Robotics (SAIR) Lab, University at Buffalo, NY 14260 · Robotics Institute, Carnegie Mellon University, PA 15213
Forget Attention: Importance-Aware Attention Is All You Need
Jun 1, 2026Suhyeong Shin, Yeongwook YangTransformer Architectures Importance
Department of Computer Engineering Kangwon National University
CRePE: Convolution-aware Relative Importance in Post-training Pruning with Efficient Search
Jun 1, 2026Cheonjun ParkUnstructured Pruning Large Language Model Compression
Hankuk University of Foreign Studies
Quotient DAGs for Off-Policy Evaluation:Forward-Flow Importance Sampling and Exact Slate Propensities
May 28, 2026Ziwen Xie, Shaowen Xiang, Hongyu He +1Policy Evaluation Importance
Shanghai Jiao Tong University · National University of Singapore
The Importance of Being Statistically Earnest: A Critical Re-evaluation of GSM-Symbolic
May 27, 2026Dominika Agnieszka Długosz, Arlindo Oliveira, Natalia Díaz-RodríguezLLM Reasoning Strategies Reasoning Skills
Instituto Superior Técnico & INESC-ID, Universidade de Lisboa, Portugal · Dept. of Computer Science and AI & DaSCI Institute, Universidad de Granada, Spain
Gaussian Rank-Based Neighborhood Degree for Graph Neural Networks in Image Classification
May 23, 2026Rafael Mendonça Duarte, Jean Roberto Ponciano, Lucas Pascotti ValemGraph Neural Networks Image Classification
Institute of Mathematics and Computer Science (ICMC) University of São Paulo (USP) São Carlos – SP – Brazil
DFSAttn: Dynamic Fine-grained Sparse Attention for Efficient Video Generation
May 22, 2026Jie Hu, Zixiang Gao, Yutong He +1Dynamic Sparse Attention Sparsity
1Peking University.
Worse than Random: The Importance of a Baseline for Unsupervised Feature Selection
May 21, 2026Muhammad Rajabinasab, Michael E. Houle, Oussama Chelly +1Feature Selection Unsupervised
University of Southern Denmark, Odense, Denmark · New Jersey Institute of Technology, Newark, NJ, United States · oratio Technologies, Tunis, Tunisia
Argo: Efficient Importance Labeling for Enterprise Email Systems
May 20, 2026Siddhant Ray, Ganesh Ananthanarayanan, Kevin Chian +5Inference Cost Importance
University of Chicago · Microsoft · University of Chicago / TensorMesh

Oct 1, 2026, cs.LGJ/K move · Enter open · S save

Don't Waste the Noise: Importance-Guided Perturbation Allocation under Joint Global and Local Constraints

Melika Shirian, Kianoosh Vadaei

Independent Researcher

Adversarial Training Quadratic Slack Penalty Importance Token Budget Allocation Structural Constraints

Open paper AskPDF arXiv

Adversarial optimization under a shared

\ell_1

budget requires deciding not only how much perturbation to use, but also where that limited budget should be spent. This allocation problem becomes particularly important when individual input coordinates are subject to local magnitude constraints, which restrict the extent to which perturbation can be concentrated on a small number of locations. We introduce an importance-guided allocation mechanism that uses a fixed clean-gradient prior to steer perturbation toward model-sensitive regions while leaving the feasible perturbation set unchanged. A centered allocation objective encourages perturbation at above-average importance locations and discourages unnecessary expenditure elsewhere, thereby redistributing rather than enlarging the available budget. Across ten robust model--dataset configurations under a common capacity-limited threat setting, the proposed method improves attack success over matched APGD- and PMA-based baselines by

2.52

17.70

percentage points. Allocation analysis shows that these gains are accompanied by substantially greater perturbation mass in high-importance regions without increased global

\ell_1

consumption. Mechanism ablations further show that centered non-uniform redistribution provides part of the benefit, while model-derived importance yields an additional improvement. These results identify perturbation allocation as a distinct and practically relevant dimension of adversarial optimization under shared-budget, locally constrained threat models.

Momentum

Related topics

Latest papers 92