AI News

DXA-Derived Skeletal Phenotypes and Hip Fracture Risk: A Backdoor-Adjusted Causal Analysis (arxiv.org)

LEAP: Supercharging LLMs for Formal Mathematics with Agentic Frameworks (arxiv.org)

RRISE: Robust Radius Inference via a Surrogate Estimator (arxiv.org)

The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (arxiv.org)

Effect of Demographic Bias on Skin Lesion Classification (arxiv.org)

Multi-Segment Attention: Enabling Efficient KV-Cache Management for Faster Large Language Model Serving (arxiv.org)

Neural Attention Search Linear: Towards Adaptive Token-Level Hybrid Attention Models (arxiv.org)

ReaLM: Residual Quantization Bridging Knowledge Graph Embeddings and Large Language Models (arxiv.org)

Discovering autonomous quantum error correction via deep reinforcement learning (arxiv.org)

GuidedBridge: Training-freely Improving Bridge Models with Prior Guidance (arxiv.org)

Learning DNF through Generalized Fourier Representations (arxiv.org)

Scaling Multi Agent Reinforcement Learning for Underwater Acoustic Tracking via Autonomous Vehicles (arxiv.org)

WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation (arxiv.org)

FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs (arxiv.org)

Typhoon: Towards an Effective Task-Specific Masking Strategy for Pre-trained Language Models (arxiv.org)

Align-KD: Distilling Cross-Modal Alignment Knowledge for Mobile Vision-Language Model Enhancement (arxiv.org)

The Agent's First Day: Benchmarking Learning, Exploration, and Scheduling in the Workplace Scenarios (arxiv.org)

Enhanced Renewable Energy Forecasting using Context-Aware Conformal Prediction (arxiv.org)

Optimizing Neuro-Fuzzy and Colonial Competition Algorithms for Skin Cancer Diagnosis in Dermatoscopic Images (arxiv.org)

Suboptimality bounds for trace-bounded SDPs enable a faster and scalable low-rank SDP solver SDPLR+ (arxiv.org)

When Should LLMs Be Less Specific? Selective Abstraction for Reliable Long-Form Text Generation (arxiv.org)

SAIL: Sound Abstract Interpreters with LLMs (arxiv.org)

Decoupled Smart Contract Audits: Lightweight LLM Framework via Distillation and Aggregation (arxiv.org)

A Single-Loop Bilevel Deep Learning Method for Optimal Control of Obstacle Problems (arxiv.org)

A Graph Foundation Model with Spectral Parsing and Prototype-Guided Spatial Propagation (arxiv.org)

Entropy Is Not Enough: Unlocking Effective Reinforcement Learning for Visual Reasoning via Vision-Anchored Token Selection (arxiv.org)

ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models (arxiv.org)

Self-Soupervision: Cooking Model Soups without Labels (arxiv.org)

Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models (arxiv.org)

Who Deserves the Reward? SHARP: Shapley Credit-based Optimization for Multi-Agent System (arxiv.org)

MIND: Multi-rationale INtegrated Discriminative Reasoning Framework for Multi-modal Large Models (arxiv.org)

DRAN: A Distribution and Relation Adaptive Network for Spatio-temporal Forecasting (arxiv.org)

FutureWeaver: Planning Test-Time Compute for Multi-Agent Systems with Modularized Collaboration (arxiv.org)

Efficient Transformer-Based Localized Patch Sampling for Choroid Plexus Segmentation in Multiple Sclerosis (arxiv.org)

Position: Adversarial ML for LLMs Is Not Making Any Progress (arxiv.org)

PINNfluence: Interpreting PINNs through Influence Functions (arxiv.org)

Annot-Mix: Learning with Noisy Class Labels from Multiple Annotators via a Mixup Extension (arxiv.org)

VLESA: Vision-Language Embodied Safety Agent for Human Activity Monitoring (arxiv.org)

Formalizing the Binding Problem (arxiv.org)

R2DN: Scalable Parameterization of Contracting and Lipschitz Recurrent Deep Networks (arxiv.org)

A Factorized Low-Rank RNN Framework for Uncovering Independent Neural Latent Dynamics and Connectivity (arxiv.org)

DiffUNet^2: Bidirectional Prediction, Probabilistic Generation and Collaborative Visual Discovery for Scientific Data (arxiv.org)

CoralBay: A Self-Supervised CT Foundation Model (arxiv.org)

Merit or networks? What decides where research is published (arxiv.org)

Visual Instruction Tuning Aligns Modalities through Abstraction (arxiv.org)

Training-Free Multi-Concept LoRA Composition with Prompt-Aware Weighting (arxiv.org)

Quadratic integrate-and-fire neurons exhibit less fragmented loss landscapes and outperform leaky integrate-and-fire neurons in spike-based gradient descent (arxiv.org)

Whom to Query for What: Adaptive Group Elicitation via Multi-Turn LLM Interactions (arxiv.org)

Expert-Aware Causal Tracing of Factual Recall in Sparse MoE Language Models (arxiv.org)

Black-box, Adaptive, Efficient, Transferable, Harmful, Applicable... Attacks Are All You Need to Break LLMs (arxiv.org)