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

推荐订阅源

Full Disclosure

Recorded Future

CERT Recently Published Vulnerability Notes

Schneier on Security

The Hacker News

CXSECURITY Database RSS Feed - CXSecurity.com

Know Your Adversary

Privacy International News Feed

Threat Intelligence Blog | Flashpoint

The Register - Security

Cisco Talos Blog

Kaspersky official blog

True Tiger Recordings

Threat Research - Cisco Blogs

Vulnerabilities – Threatpost

Palo Alto Networks Blog

The Exploit Database - CXSecurity.com

Cyber Security Advisories - MS-ISAC

Microsoft Azure Blog

Cybersecurity and Infrastructure Security Agency CISA

Tor Project blog

Proofpoint News Feed

Fox-IT International blog

Fortinet All Blogs

Privacy & Cybersecurity Law Blog

OSCHINA 社区最新新闻

博客园 - 叶小钗

Tailwind CSS Blog

Netflix TechBlog - Medium

IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog

博客园_首页

Hackread – Cybersecurity News, Data Breaches, AI and More

Darknet – Hacking Tools, Hacker News & Cyber Security

cs.LG updates on arXiv.org

Recursive Flow Matching Model Merging on Loss Landscape: A Geometry Perspective JLT: Clean-Latent Prediction in Latent Diffusion Transformers Auditing and Fixing Economic Validity in Tabular Foundation Models for Discrete Choice Classification and detection of multiple UAVs using rational Gaussian wavelet neural networks When Correct Demonstrations Hurt: Rethinking the Role of Exemplars in In-Context Learning Dense2MoE: Pushing the Pareto Frontier of On-Device LLMs via Unified Pruning and Upcycling DDGAD: Trajectory Dynamics for Diffusion-Based Graph Anomaly Detection On the Role of Inductive Bias in Time-Series Pretraining: A Case Study in Learning Generalizable Representations for Clinical Time Series PIDM-DP: Physics-Informed Diffusion with Dormand-Prince Integration for Chaotic System Identification and State Reconstruction across Multiple Dynamical Regimes Balancing Plasticity and Stability with Fast and Slow Successor Features Rotation-Invariant Spherical Watermarking via Third-Order SO(3) Representation Coupling Max-Window Scale Estimation for Near-Lossless HiF8 W8A8 Quantization-Aware Training Few-shot Cross-country Generalization of Tabular Machine Learning and Foundation Models for Childhood Anemia Prediction under Distribution Shift HRVConformer: Neonatal Hypoxic-Ischemic Encephalopathy Classification from the Heart Rate signals APEX: Amplitude Anchors and Phase Priors for Target-Scarce Higher-Frequency Wave Prediction Aligning Few-Step Generative Models by Amortizing Sample-based Variational Inference Bridging Classification and Reconstruction: Cooperative Time Series Anomaly Detection Stabilizing Recurrent Dynamics for Test-Time Scalable Latent Reasoning in Looped Language Models TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models Variational Inference for Evidential Deep Learning CSV-ViT: A Vision Transformer with the Variable-sized Cortical Supervertices for Detection of Alzheimer's Disease Pathologies Open-Weight LLM Fine-Tuning Defenses are Susceptible to Simple Attacks FM-fMRI: Event Conditioned Flow Matching for Rest-to-Task fMRI Time-Series Synthesis QAM-W: Joint 2D Codebook Quantization for LLM Weights via Hadamard Rotation and Activation-Aware Scaling SilIF: Silhouette-Augmented Isolation Forest for Unsupervised Transaction Fraud Detection The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works TrackRef3D: Multi-View Consistent Track-then-Label for Open-World Referring Segmentation in 3D Gaussian Splatting A PAC-Bayesian View of Generalisation for Physics-Informed Machine Learning Towards Controllable Image Generation through Representation-Conditioned Diffusion Models Linear and Neural Dueling Bandits with Delayed Feedback Adversarial Training for Robust Coverage Network under Worst-case Facility Losses When Rule Violations Are Rare: Chimera Training for Logical Anomaly Detection WINDQuant: Weight-Informed Neural Decision-Making for Global Mixed-Precision LLM Quantization The Constraint Tax: Measuring Validity-Correctness Tradeoffs in Structured Outputs for Small Language Models RT-Lynx: Putting the GEMM Sparsity In a Right Way for Diffusion Models Neural Bayesian Sequential Routing Self-Improvement Imitation with Biologically Guided Search for Protein Design Under Oracle Budgets Bilevel Optimization over Saddle Points of Zero-Sum Markov Games Provably Communication-Efficient and Privacy-Preserving Federated Graph Neural Networks MTL-FNO: A Lightweight Multi-Task Fourier Neural Operator for Sparse Field Reconstruction Aperiodic and Low-Frequency Spectral Bias in Reconstruction based EEG Foundation Models Reparametrizing Shampoo and SOAP for Subspace Basis Updates and BFloat16 Storage Unified Neural Scaling Laws MuCon: Clipped Muon Updates for LLM Training LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding GEM: Geometric Entropy Mixing for Optimal LLM Data Curation MULTISEISMO: A Multimodal Seismic Dataset and Model for Cross-Modal Seismic Understanding A Fast and Generic Energy-Shifting Transformer for Hybrid Monte Carlo Radiotherapy Calculation Benchmarking Convolutional, Transformer, Hybrid, and Vision Language Models for Multi Disease Retinal Screening BioFact-MoE: Biologically Factorized Mixture of Experts for Vision-Language Prognostic Modeling in Hepatocellular Carcinoma PRISM: Position-encoded Regressive Inverse Spectral Model for Multilayer Thin-Film Design Time Series Causal Discovery via Context-Conditioned and Causality-Augmented Pretraining Energy-Gated Attention and Wavelet Positional Encoding: Complementary Inductive Biases for Transformer Attention Semigroup Consistency as a Diagnostic for Learned Physics Simulators Planning Neural Dynamics with Lie Group Embedding through Supervised Projective Manifold Learning Function-Valued Causal Influence in Nonlinear Time Series Extra-Merge: Tracing the Rank-1 Subspace of Model Merging in Language Model Pre-Training Online Learning on Hidden-Convex Losses via Algorithmic Equivalence: Optimal Regret, Geometric Barrier, and Bandit Feedback Quantized Keys Steal Attention: Bias Correction for KV-Cache Compression in Video Diffusion Curriculum Learning for Safety Alignment SIKA-GP: Accelerating Gaussian Process Inference with Sparse Inducing Kernel Approximations for Bayesian Deep Learning Personalized Generative Models for Contextual Debiasing Geometry-Aware Contrastive Learning for Few-Shot Automatic Modulation Recognition Beyond Holistic Models: Systematic Component-level Benchmarking of Deep Multivariate Time-Series Forecasting Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learning Distribution-Aware Conformal Prediction: A Framework for generating efficient prediction intervals for time series On the Error-Correcting Effects of Stochasticity in Discrete Diffusion On the Push-Based Asynchronous Federated Learning: A Bias-Correction Aggregation Approach More Expressive Feedforward Layers: Part I. Token-Adaptive Mixing of Activations Separate Aggregation of Split Network for Personalized Federated Learning SL-BiLEM: Structured Learnable Behavior-in-the-Loop Epidemic Modeling for Forecasting and Policy Evaluation GAC: Noise-Aware Adaptive Mixing for Hybrid SFT-RL Post-Training Localizing Memorized Regions in Diffusion Models via Coordinate-Wise Curvature Differences A Hybrid Vision-Language Architecture for Automated Defect Reasoning and Report Generation in Industrial Inspection FoundObj: Self-supervised Foundation Models as Rewards for Label-free 3D Object Segmentation Ratio-Variance Regularized Policy Optimization Focal Reward: Balanced Reinforcement Learning under Rubric-Based Rewards Scaling World-Model Reinforcement Learning Through Diffusion Policy Optimization Is an Image Also Worth 16x16=256 Superpixels? A Framework for Attentional Image Classification From Privacy to Generalization: Linear Max-Information Bounds for DP-SGD Diffuse to Detect: Generative Diffusion Models for Unsupervised IC Anomaly Detection Modeling Dynamic Mixtures of Time-Delay Systems from Streaming Time Series Stateful Inference for Low-Latency Multi-Agent Tool Calling When Does Deep RL Beat Calibrated Baselines? A Benchmark Study on Adaptive Resource Control Two-Parameter Flows for Learning Population Dynamics of Physical Systems MechRL: Reinforcement Learning Agents Perform Circuit Discovery for Mechanistic Interpretability InfoQuant: Shaping Activation Distributions for Low-Bit LLM Quantization Pretrained Approximators for Low-Thrust Trajectory Cost and Reachability Co-folding model guided by structural proteomics Towards Generalization-Oriented Models for Vehicle Routing Problems with Mixture-of-Experts AirCast-SR: A Foundation Model for Kilometer-Scale Atmospheric Super-Resolution via Latent Consistency Diffusion ARBITER: Reasoning Trajectory Basins and Majority Vote Failures in Test-Time Sampling Beyond Pairwise Preferences: Listwise Reward-Aware Alignment for Diffusion Models Image Feature Fusion-based Federated Client Unlearning (FCU) Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training PILOT: A Data-Free Continual Learning Approach for Real-Time Semantic Segmentation via Boundary Guidance Near-Optimal Regret in Adversarial Kernel Bandits Amortized Factor Inference Networks for Posterior Inference Dynamic Link Prediction with Temporally Enhanced Signed Graph Neural Networks

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

Jimyung Hong · 2026-05-05 · via cs.LG updates on arXiv.org

此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。原文来自 — 版权归原作者所有。