AI News

Veille quotidienne — 14/09/2026

Who Gets to Define the Rules for AI? (cohere.com)

Inverse-Square Law (blog.coredump.cx)

OPINION: NJ should embrace data centers while limiting impacts - NJBIZ (njbiz.com)

Robust Prototypical Networks for Few-Shot Sensor Fault Diagnosis (arxiv.org)

Where Decoder Cosine Similarity Fails for SAE Feature Flow Discovery (arxiv.org)

Quality-Constrained Routing over a Fixed Pool of Quantized Mixture-of-Experts Instances (arxiv.org)

TokenMapper: A Step Toward Interoperable Speech Token Translation (arxiv.org)

SCOPE-OPSD: Fisher-Conditioned Privileged Subspaces for On-Policy Self-Distillation (arxiv.org)

Clustering-Based Balanced Sampling and Allocation with Data Parallelism for High-Performance Fine-Tuning (arxiv.org)

Poisson-Corrector Complexity Bounds for Moreau--Yosida Unadjusted Langevin Sampling (arxiv.org)

SIMS: Scale-Invariant Merit-Function-Based Scalarization for Multi-Task Learning (arxiv.org)

Correlation-Guided Fast Machine Unlearning via Hessian Analysis (arxiv.org)

Geometric-to-Semantic Spherical Transfer Learning for Cortical Sulci Labeling (arxiv.org)

Explaining Time Series Forecasting with Horizon-Resolved Attribution (arxiv.org)

Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner (arxiv.org)

ProactiveBench: Can Streaming Video Models Really Interact Like Humans? (arxiv.org)

SIFPBPNet: A Dual-Path Network for Wearable and Cuffless Blood Pressure Estimation via Individualized Steady-state Representation (arxiv.org)

Write on Paper and Get the Online Digital Trace:\newline A New Era for Handwriting (arxiv.org)

InRTL: Effective Intra-Inter Interaction Learning for Relational Tables (arxiv.org)

Physics-Guided Synthetic High-Frequency Ultrasound Generation for Skin Layer Segmentation (arxiv.org)

Optimizing for the decision not the prediction: an exploration of Smooth Net Benefit as a training objective (arxiv.org)

Curriculum-Based Adversarial Heterogeneous Agent Reinforcement Learning for Autonomous Quad-Copter Landing in Maritime Settings (arxiv.org)

VertiFuseX: Generalizable Financial Forecasting via Multi-Stream Temporal Fusion (arxiv.org)

RunningTensor: Generalizing Linear Attention to Higher-Order Recurrent States (arxiv.org)

GenOR-Twin: A Semantic Middleware for Integrating Operational Discourse with Mathematical Optimization (arxiv.org)

Learning the Geometry of Collider Events with Metric-Aware Deep Sets (arxiv.org)

What an odour descriptor corpus can and cannot measure: valence, attenuation, and the ceiling of the public record (arxiv.org)

Code-to-Harness: Distilling Black-Box Optimizers from Self-Play (arxiv.org)

AnyView: Synthesizing Any Novel View in Dynamic Scenes (arxiv.org)

PACEvolve: Enabling Progress-Aware Consistent Evolution (arxiv.org)

Deep learning methods for inverse problems using connections between proximal operators and Hamilton-Jacobi equations (arxiv.org)

HoliBench: A Cross-Platform Benchmarking and Deployment Toolkit for Foundation Models in CPS-IoT Applications (arxiv.org)

Machine Learning-Based Classification of Jhana Advanced Concentrative Absorption Meditation Using 7 Tesla Functional Magnetic Resonance Imaging (arxiv.org)

In-Hospital Stroke Risk-State Classification from PPG-Derived Hemodynamic Features (arxiv.org)

Benford's Law as a Distributional Prior for Post-Training Quantization of Large Language Models (arxiv.org)

Large Distant Gradients Need Not Be Reliable: reliability-weighted credit assignment for long-horizon autoregressive forecasting (arxiv.org)

Quantifying the Value of Privileged Information Using a PAC-Bayesian Approach (arxiv.org)

Behavior Quotient Learning for Low-Rank Adaptation of LLM Agents (arxiv.org)

Physical-State-Guided Diffusion Sampling for Full-Waveform Inversion (arxiv.org)

Hidden in Rounds: Predicting the Time Cost of 802.11 Contention in Federated Learning (arxiv.org)

Offline Reinforcement Learning for Wind Farm Control: A Wind Tunnel Study under Dynamic Wind Directions (arxiv.org)

A Large-Scale AIS Dataset from Finnish Water (arxiv.org)

Information-Induced Training Geometry: Exact Reduction, Canonical Completion, and Structured Expressivity (arxiv.org)

Dual-guided Hierarchical Edge Localization for Large-scale Optimal Transport Across Dimensions (arxiv.org)

Attention Quantization for Tabular Foundation Models (arxiv.org)

Groupoid-Based Internal State Representations for Reinforcement Learning with Local Symmetries (arxiv.org)

FRIST: FMRI Representation Informed Shared-space Training Improves EEG-only Individual-Finger BCI Decoding (arxiv.org)

DynSHAP: Towards Explainable Dynamic Survival Analysis (arxiv.org)

A Dataset and Benchmarks for Atrial Fibrillation Detection from Electrocardiograms of Intensive Care Unit Patients (arxiv.org)