AI News

Kaininja: Extending Native 3D Generators to the Part Level (arxiv.org)

CIDERS: Cloud-Edge LLM Collaborative Learning via Accelerating Personalized Bilevel Optimization (arxiv.org)

Benchmarking Intra-Patient 3D Deformable Multimodal Image Registration (arxiv.org)

Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models (arxiv.org)

Scalability and Performance Evaluation of Federated Learning Frameworks: A Comparative Analysis (arxiv.org)

More Than Just Access: Generative AI as Communication Intermediary for Blind and Low-Vision Users (arxiv.org)

Bench2Dex: Benchmarking Visuo-Tactile Bimanual Dexterous Manipulation Across Dexterous Hands (arxiv.org)

Look Before You Leap: Factual Decoding with Internal Attribution Signals (arxiv.org)

Event-Native Symbolic-Temporal Spike Encoding Framework for Heterogeneous Cyber Streams (arxiv.org)

CiteGuard-RAG: A Validation-Centered AI System for Evidence-Grounded Question Answering (arxiv.org)

Before You Poll with LLMs: A Deliberative Diagnostic Framework (arxiv.org)

Anatomical Grounding and Leakage-Aware Multimodal Contrastive Learning for Alzheimer's Disease Classification from Structural MRI (arxiv.org)

Vulnerability Localization Benchmark: Measuring Agentic Security Analysis at Repository Scale (arxiv.org)

Estimating Uncertain Spatial Relationships in Robotics (arxiv.org)

FICAug: Feature-Informed Clustering and Augmentation for Facial-Expression-Based Parkinson's Disease Screening (arxiv.org)

One Size Does Not Fit All: Setting Inference Depth from the Questions a Deployment Actually Asks (arxiv.org)

Sanity Checking Causal Representation Learning on a Simple Real-World System (arxiv.org)

L-Lipschitz Gershgorin ResNet Network (arxiv.org)

WaveHiTS: Wavelet-Enhanced Hierarchical Time Series Modeling for Wind Direction Nowcasting in Eastern Inner Mongolia (arxiv.org)

Neural Network Operator-Based Fractal Approximation: Smoothness Preservation and Convergence Analysis (arxiv.org)

Dissociating performance from compositional feature learning (arxiv.org)

Tensorization is a powerful but underexplored tool for compression and interpretability of neural networks (arxiv.org)

IsingFormer: Augmenting Parallel Tempering With Learned Proposals (arxiv.org)

MuPlon: Multi-Path Causal Optimization for Claim Verification through Controlling Confounding (arxiv.org)

Convergence, design and training of continuous-time dropout as a random batch method (arxiv.org)

Isolation-based Spherical Ensemble Representations for Tabular Anomaly Detection (arxiv.org)

Mesh-based Super-resolution of Multiscale Detonation Flows with Graph Transformers (arxiv.org)

CLIMATEAGENT: Multi-Agent Orchestration for Complex Climate Data Science Workflows (arxiv.org)

Learning Steerable Clarification Policies with Collaborative Self-play (arxiv.org)

Architecture--Optimization Co-Design for Physics-Informed Neural Networks via Layer-wise Coordinate Adaptation and Gradient Conflict Resolution (arxiv.org)

Hybrid Feedback-Guided Optimal Learning for Wireless Interactive Panoramic Scene Delivery (arxiv.org)

Dreaming in Code for Curriculum Learning in Open-Ended Worlds (arxiv.org)

Disparate Impact in Synthetic Data Generation (arxiv.org)

UniRank: Unified Rank Allocation for Low-Rank LLM Compression (arxiv.org)

AbFlow : End-to-end Paratope-Centric Antibody Design by Interaction Enhanced Flow Matching (arxiv.org)

Learning Human-Like Badminton Skills for Humanoid Robots (arxiv.org)

Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges (arxiv.org)

SDUs DAISY: A Benchmark for Danish Culture (arxiv.org)

GRADE: Graph Representation of LLM Agent Dependency and Execution (arxiv.org)

Concept drift mitigation through community and spectral graph analysis for the detection of cyberattacks in network traffic (arxiv.org)

CHAI for LLMs: Improving Code-Mixed Translation in Large Language Models through Reinforcement Learning with AI Feedback (arxiv.org)

Programs as Singularities (arxiv.org)

Exact Spin Elimination for Quadratic and k-Local Ising Optimization (arxiv.org)

Efficient On-Device Agents via Adaptive Context Management (arxiv.org)

Hallucination in Multimodal Foundation Models: A Survey on Causes, Corrections, and Evaluations (arxiv.org)

DeepFeature: LLM-Empowered Context-aware Feature Generation for Wearable Biosignals (arxiv.org)

Privileged observations enable rapid and reliable policy discovery directly in the physical world (arxiv.org)

MCPAgentBench: A Real-world Task Benchmark for Evaluating LLM Agent MCP Tool Use (arxiv.org)

Towards a Mechanistic Understanding of Propositional Logical Reasoning in Large Language Models (arxiv.org)

CLEAR: Context Augmentation from Contrastive Learning of Experience via Agentic Reflection (arxiv.org)