AI News

A Lightweight Hybrid MLP-Based Framework for Real-Time Phishing URL Detection Using Structural URL Features (arxiv.org)

Prototypicality Bias Reveals Blindspots in Multimodal Evaluation Metrics (arxiv.org)

Efficient Exploration for Iterative Nash Preference Optimization (arxiv.org)

OP-LoRA: The Blessing of Dimensionality (arxiv.org)

Multilinguality of Large Language Models From a Structural Perspective (arxiv.org)

Deft Scheduling of Dynamic Cloud Workflows with Varying Deadlines via Mixture-of-Experts (arxiv.org)

Application of Algorithms in Energy-Efficient Design Platforms for Green Building (arxiv.org)

MENTIS: What Belief Changes Under Alignment? Measuring Multi-Scale Latent Torsion in Language Models (arxiv.org)

Local Preferential Bayesian Optimization (arxiv.org)

Universal One-third Time Scaling in Learning Peaked Distributions (arxiv.org)

Agent Operating Systems (AOS): Integrating Agentic Control Planes into, and Beyond, Traditional Operating Systems (arxiv.org)

LLMs Need Encoders for Semantic IDs Too (arxiv.org)

The role of class encoding in neural collapse (arxiv.org)

MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation (arxiv.org)

Real2SAM2Real: Generative 3D Caches as Complementary Context for Video Diffusion (arxiv.org)

Knowing Isn't Understanding: Re-grounding Generative Proactivity with Epistemic and Behavioral Insight (arxiv.org)

Drifting Preference Optimization for One-Step Generative Models (arxiv.org)

Isolating LLM Lexical Bias: A Curation-Free Triangulated Metric for Preference-Stage Learning (arxiv.org)

Trajectory Data Suffices for Statistically Efficient Policy Evaluation in Fixed-Horizon Offline RL with Linear $q^\pi$-Realizability and Concentrability (arxiv.org)

Generative AI and Digital Ecosystem Resilience: A Proactive Lifecycle-Based Survey (arxiv.org)

Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance (arxiv.org)

Safeguarded Stochastic Polyak Step Sizes for Non-smooth Optimization: Robust Performance Without Small (Sub)Gradients (arxiv.org)

DRL-Based Pose Control for Double-Ackermann Robots Under Actuation Uncertainties (arxiv.org)

Guidance for Low-Level Perceptual Editing in Unconditional Diffusion Models (arxiv.org)

3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code (arxiv.org)

Neural Network Compression by Approximate Differential Equivalence (arxiv.org)

Entropy Minimization without Model Collapse: Mitigating Prediction Bias in Medical Imaging (arxiv.org)

MemPro: Agentic Memory Systems as Evolvable Programs (arxiv.org)

Hyperbolic and Evidence-Prioritized Experts for Large Vision-Language Models (arxiv.org)

Verifying Meta-Awareness via Predictive Rewards in Reasoning Models (arxiv.org)

MGRegBench: A Novel Benchmark Dataset with Anatomical Landmarks for Mammography Image Registration (arxiv.org)

Effects of Varying LLM Access on Essay Writing Behavior (arxiv.org)

A Multi-AI-agent Framework Enabling End-to-end Finite Element Analysis for Solid Mechanics Problems (arxiv.org)

PaintBench: Deterministic Evaluation of Precise Visual Editing (arxiv.org)

Don't Read Everything: A Curvature-Conditioned Query for Linear Attention (arxiv.org)

Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation (arxiv.org)

Safe-Subspace Pseudo-Label Refinement for Source-Free Graph Domain Adaptation (arxiv.org)

Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs (arxiv.org)

A Primer in Post-Training Reasoning Data: What We Know About How It Works (arxiv.org)

Pause and Think: A Dataset and Benchmark for Video-Grounded Assistive Action Suggestion (arxiv.org)

From "Weak" Signals to Strong Models: Preference Delta Aggregation with LoRA Merging (arxiv.org)

GC-MoE: Genomics-Guided Cell-Type-Specific Mixture of Experts for Histology-Based Single-Cell Spatial Transcriptomics (arxiv.org)

Permissive Safety Through Trusted Inference: Verifiable Belief-Space Neural Safety Filters for Assured Interactive Robotics (arxiv.org)

Fixed-Mean Gaussian Processes for Post-hoc Bayesian Deep Learning (arxiv.org)

LastAct: Trajectory-Guided Latest-Activity Localization for Real-Time Smart-Home Activity Recognition (arxiv.org)

Latent Reasoning in TRMs is Secretly a Policy Improvement Operator (arxiv.org)

Emergent Ordinal Geometry in Transformers Trained on Local Comparisons (arxiv.org)

CardioLens: Revealing the Clinical Reality Gap of MLLMs via Multi-Sequence Cardiac MRI Evaluations (arxiv.org)

Structure-Guided Adaptive Propagation for Protein-Protein Interaction Site Prediction (arxiv.org)

KISS: Keeping it Simple and Slotted when Learning to Communicate over Wireless (arxiv.org)