AI News

The role of locality and weight sharing in convolutional neural networks (arxiv.org)

Mitigating Model Collapse through Dreaming Learning (arxiv.org)

Mastering Model Fit: Applications of the InterModel Vigorish for Interpretable Machine Learning (arxiv.org)

How Do DiTs Denoise? An Aggregator-Predictor View of Attention and MLPs (arxiv.org)

Asymmetric Likelihood-Free Hypothesis Testing (arxiv.org)

Post-hoc Epistemic Uncertainty for Vision-Language-Action Policies (arxiv.org)

Fragment-Level Graph Encoding for Interpretable Molecular Property Prediction (arxiv.org)

Private Online Prediction from Experts with Small Losses (arxiv.org)

BRIDGE: An Acceptance Gate to Block False Improvements in Model Updates, with a Retrospective Clinical Evaluation (arxiv.org)

Generative Modeling via Nonlinear Schr\"odinger Bridges and Branching Diffusions (arxiv.org)

Density-Level Backward Error for Low-NFE Generative Flows (arxiv.org)

One Act, Many Masks: Effect-Marginal Policy Optimization for Long-Horizon Agents (arxiv.org)

The Predictive--Optimization Gap in GuanDan Self-Play (arxiv.org)

Cascading Bandits with Noisy Observations: Algorithms and Regret Bounds (arxiv.org)

Agent Behavior Maps: Locating Agents by What They Would Do, Without Running Them (arxiv.org)

GaP-MoE : Explicit Expert Specialization in Mixture-of-Experts via Orthogonal Gradient Accumulation and Projection (arxiv.org)

ExpEvolve: Structuring and Reusing Experience for Open-Ended Discovery (arxiv.org)

Approximate Solomonoff Induction: On the Robustness of Learned Universal Predictors (arxiv.org)

Better Policies, Worse Verifier Audits (arxiv.org)

Curvature-Scaled Sharpness-Aware Minimization: Convergence Beyond Smoothness (arxiv.org)

A Jump-Modified Wasserstein Gradient Flow for Maximum Mean Discrepancy with Energy Kernels (arxiv.org)

Online Open-World Wafer Defect Discovery using Hyperdimensional Computing (arxiv.org)

From Common Information to Canonical Correlation Analysis and Beyond: Bridging Information Theory, Signal Processing, and Machine Learning (arxiv.org)

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks (arxiv.org)

Backtracking Direct, Indirect and Total Effects (arxiv.org)

Actually, All Learnable Distribution Classes Are Privately Learnable (arxiv.org)

ReAPO: Retry-Aware Policy Optimization for Efficient Exploration in LLM Reasoning (arxiv.org)

The Price of Current-Effect Inference in Nonstationary Multi-Armed Bandits (arxiv.org)

ART-MPG: Autoregressive R/T Manifold Path Guiding (arxiv.org)

Covariate-assisted graph matching (arxiv.org)

Silent Alignment in Two-Layer Linear Networks: A High-Dimensional Perspective (arxiv.org)

Lifting PCA in the Wasserstein Space (arxiv.org)

Near Minimax Rate for Quantile Distributional Reinforcement Learning (arxiv.org)

Residual-Based Causal Discovery via Pretrained In-Context Quantile Regression (arxiv.org)

A unified approach for estimation of unnormalized models via Stein operators (arxiv.org)

When Thoughts Become Memory: Unifying In-Context Learning and Fine-Tuning (arxiv.org)

Schr\"odinger-Bass Bridge Barycenter Models for Measure-Valued Data (arxiv.org)

Internal Guidance as a Sampling Method for Tilted Diffusion Models (arxiv.org)

Is In-Context Reinforcement Learning A Posterior Sampler? An Empirical Study (arxiv.org)

Depth Separation for Neural Networks under Gaussian Measure (arxiv.org)

A Reverse-Gold Theorem: Positive-Data Identification by Substrate-Bounded Contrastive Learners (arxiv.org)

Borrowing Strength Across Stores: Transfer Learning for Choice-Based Revenue Management (arxiv.org)

NVIDIA Nsight Python Tutorial: How to Profile GPU Kernels (www.marktechpost.com)

I can't believe PowerShell is good (www.spacebar.news)

In Honduras, hard hit by drought, authorities hope to make it rain using contentious tactic (wtop.com)

Startup Claims To Use AI-Created Viruses To Treat a Patient (science.slashdot.org)

Franklin County holds informational panel on data center development - WSLS (www.wsls.com)

Flapping Airplanes (flappingairplanes.com)

The recently proposed O(n log n) integer multiplication algorithm [pdf] (web.maths.unsw.edu.au)

Hackers Exploited the PHP Vulnerabilities Days Before They Were Patched (www.wired.com)