Content Feed

The latest content from across the network

Ranked Latest Top this week
[2604.02659] Low-Rank Compression of Pretrained Models via Randomized Subspace Iteration
Machine Learning

[2604.02659] Low-Rank Compression of Pretrained Models via Randomized Subspace Iteration

Abstract page for arXiv paper 2604.02659: Low-Rank Compression of Pretrained Models via Randomized Subspace Iteration

arXiv - AI · 3 min ·
[2604.02653] Product-Stability: Provable Convergence for Gradient Descent on the Edge of Stability
Machine Learning

[2604.02653] Product-Stability: Provable Convergence for Gradient Descent on the Edge of Stability

Abstract page for arXiv paper 2604.02653: Product-Stability: Provable Convergence for Gradient Descent on the Edge of Stability

arXiv - Machine Learning · 3 min ·
[2604.02652] Generalization Limits of Reinforcement Learning Alignment
Llms

[2604.02652] Generalization Limits of Reinforcement Learning Alignment

Abstract page for arXiv paper 2604.02652: Generalization Limits of Reinforcement Learning Alignment

arXiv - AI · 3 min ·
[2604.02651] Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training
Machine Learning

[2604.02651] Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training

Abstract page for arXiv paper 2604.02651: Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training

arXiv - AI · 4 min ·
[2604.02644] Conditional Sampling via Wasserstein Autoencoders and Triangular Transport
Machine Learning

[2604.02644] Conditional Sampling via Wasserstein Autoencoders and Triangular Transport

Abstract page for arXiv paper 2604.02644: Conditional Sampling via Wasserstein Autoencoders and Triangular Transport

arXiv - Machine Learning · 3 min ·
[2604.02638] AXELRAM: Quantize Once, Never Dequantize

[2604.02638] AXELRAM: Quantize Once, Never Dequantize

Abstract page for arXiv paper 2604.02638: AXELRAM: Quantize Once, Never Dequantize

arXiv - Machine Learning · 3 min ·
[2604.02633] Analytic Drift Resister for Non-Exemplar Continual Graph Learning

[2604.02633] Analytic Drift Resister for Non-Exemplar Continual Graph Learning

Abstract page for arXiv paper 2604.02633: Analytic Drift Resister for Non-Exemplar Continual Graph Learning

arXiv - AI · 3 min ·
[2604.02615] Complex-Valued GNNs for Distributed Basis-Invariant Control of Planar Systems
Machine Learning

[2604.02615] Complex-Valued GNNs for Distributed Basis-Invariant Control of Planar Systems

Abstract page for arXiv paper 2604.02615: Complex-Valued GNNs for Distributed Basis-Invariant Control of Planar Systems

arXiv - Machine Learning · 3 min ·
[2604.02608] Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens
Llms

[2604.02608] Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens

Abstract page for arXiv paper 2604.02608: Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens

arXiv - Machine Learning · 4 min ·
[2604.02601] WGFINNs: Weak formulation-based GENERIC formalism informed neural networks'
Machine Learning

[2604.02601] WGFINNs: Weak formulation-based GENERIC formalism informed neural networks'

Abstract page for arXiv paper 2604.02601: WGFINNs: Weak formulation-based GENERIC formalism informed neural networks'

arXiv - Machine Learning · 4 min ·
[2604.02580] VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generation
Machine Learning

[2604.02580] VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generation

Abstract page for arXiv paper 2604.02580: VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generation

arXiv - Machine Learning · 3 min ·
[2604.02577] ROMAN: A Multiscale Routing Operator for Convolutional Time Series Models
Machine Learning

[2604.02577] ROMAN: A Multiscale Routing Operator for Convolutional Time Series Models

Abstract page for arXiv paper 2604.02577: ROMAN: A Multiscale Routing Operator for Convolutional Time Series Models

arXiv - Machine Learning · 4 min ·
[2604.02558] Communication-Efficient Distributed Learning with Differential Privacy
Machine Learning

[2604.02558] Communication-Efficient Distributed Learning with Differential Privacy

Abstract page for arXiv paper 2604.02558: Communication-Efficient Distributed Learning with Differential Privacy

arXiv - Machine Learning · 3 min ·
[2604.02556] Fast NF4 Dequantization Kernels for Large Language Model Inference
Llms

[2604.02556] Fast NF4 Dequantization Kernels for Large Language Model Inference

Abstract page for arXiv paper 2604.02556: Fast NF4 Dequantization Kernels for Large Language Model Inference

arXiv - Machine Learning · 3 min ·
[2604.02535] A Spectral Framework for Multi-Scale Nonlinear Dimensionality Reduction

[2604.02535] A Spectral Framework for Multi-Scale Nonlinear Dimensionality Reduction

Abstract page for arXiv paper 2604.02535: A Spectral Framework for Multi-Scale Nonlinear Dimensionality Reduction

arXiv - Machine Learning · 3 min ·
[2604.02527] Jump Start or False Start? A Theoretical and Empirical Evaluation of LLM-initialized Bandits
Llms

[2604.02527] Jump Start or False Start? A Theoretical and Empirical Evaluation of LLM-initialized Bandits

Abstract page for arXiv paper 2604.02527: Jump Start or False Start? A Theoretical and Empirical Evaluation of LLM-initialized Bandits

arXiv - AI · 4 min ·
[2604.02525] AdaHOP: Fast and Accurate Low-Precision Training via Outlier-Pattern-Aware Rotation
Llms

[2604.02525] AdaHOP: Fast and Accurate Low-Precision Training via Outlier-Pattern-Aware Rotation

Abstract page for arXiv paper 2604.02525: AdaHOP: Fast and Accurate Low-Precision Training via Outlier-Pattern-Aware Rotation

arXiv - Machine Learning · 4 min ·
[2604.02511] Re-analysis of the Human Transcription Factor Atlas Recovers TF-Specific Signatures from Pooled Single-Cell Screens with Missing Controls
Data Science

[2604.02511] Re-analysis of the Human Transcription Factor Atlas Recovers TF-Specific Signatures from Pooled Single-Cell Screens with Missing Controls

Abstract page for arXiv paper 2604.02511: Re-analysis of the Human Transcription Factor Atlas Recovers TF-Specific Signatures from Pooled...

arXiv - Machine Learning · 4 min ·
[2604.02488] Causal-Audit: A Framework for Risk Assessment of Assumption Violations in Time-Series Causal Discovery
Ai Startups

[2604.02488] Causal-Audit: A Framework for Risk Assessment of Assumption Violations in Time-Series Causal Discovery

Abstract page for arXiv paper 2604.02488: Causal-Audit: A Framework for Risk Assessment of Assumption Violations in Time-Series Causal Di...

arXiv - Machine Learning · 4 min ·
[2604.02482] SEDGE: Structural Extrapolated Data Generation

[2604.02482] SEDGE: Structural Extrapolated Data Generation

Abstract page for arXiv paper 2604.02482: SEDGE: Structural Extrapolated Data Generation

arXiv - Machine Learning · 3 min ·