AI News

Introducing SINFONIA: Symplectic, slimplectic and Magnusian (Neural) Flows for Orbital Numerical Integration and Acceleration (arxiv.org)

Beyond .WAV: Design and Software Verification of VocalCap, a Traceable Browser-Based Audio Capture System for Vocal Biomarker Research (arxiv.org)

What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation (arxiv.org)

What is Smoothness? (arxiv.org)

Counterfactual Fairness Audits of Multi-Step Clinical LLM Agents Require a Measured Per-Action Instability Floor (arxiv.org)

SWIM: Student Writing Simulation via Proficiency-Conditioned Generation (arxiv.org)

VoxReason: Listener-Free Evaluation of Source-Grounded Speech Planning Before Synthesis (arxiv.org)

MemoryLACE: Memory Lifecycle-Aware Consolidation and Evidence Retrieval (arxiv.org)

Generative Nested Sampling of Atomistic Thermodynamic Landscapes (arxiv.org)

Coupled Tensor-Tensor Completion Method with Applications in Drug Repurposing (arxiv.org)

Who Speaks for the Pruned? Visual Token Pruning as Coverage Optimization (arxiv.org)

BASP: Communication-Efficient Batch-Aware Sequence Parallelism for LLM Training (arxiv.org)

RACE-AIMC: Selective Inference for Heterogeneous Analog In-Memory Accelerators at the Edge (arxiv.org)

Feasible but Not Safe: Constraint Violations and Report-Channel Attacks in Learned Cell-Free ISAC Association (arxiv.org)

Sensing Which Modality Matters: Evidence-Gated Regularization for Robust VLA Policies (arxiv.org)

CRAW: Codec Robust Audio Watermarking (arxiv.org)

Beyond Blur: A Semantic Tri-view Pipeline for Teledermatology Gradability via Skin Micro-relief (arxiv.org)

Position: Unlabeled IS NOT Equal to No Human Supervision in Visual Learning (arxiv.org)

Differentially private federated learning with Byzantine-robust aggregation: A cross-domain framework for secure model training in banking and healthcare systems (arxiv.org)

IDSPACE: A Novel Document Generator for Reliable Evaluation of Digital Identity Verification Systems [Extended Technical Report] (arxiv.org)

PatchBench: Evaluating AI Agents for Vulnerability Patching (arxiv.org)

LDC: Learning to Generate Research Idea with Dynamic Control (arxiv.org)

Why Gated DeltaNet Survives 4-Bit Quantization: NVFP4 W4A4 for the Recurrent Half of a Hybrid 27B LLM (arxiv.org)

Data Market Design through Deep Learning (arxiv.org)

HOMURA: Taming the Sand-Glass for Time-Constrained LLM Translation via Reinforcement Learning (arxiv.org)

Evolving Excellence: Automated Optimization of LLM-based Agents (arxiv.org)

Advances in Machine Learning for Directed Evolution: A Five-Year Retrospective (arxiv.org)

Discovering High Level Patterns from Simulation Traces (arxiv.org)

The Dually Flat Geometry of Planning as Inference (arxiv.org)

Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judge (arxiv.org)

Not All Preferences Deserve Gradients: Understanding Gradient Utility in Offline Reasoning Alignment (arxiv.org)

Complete Identification of Deep ReLU Networks through {\L}ukasiewicz Logic (arxiv.org)

ESPO: Error-Structured Prompt Optimization via Diagnose, Diversify, and Stabilize (arxiv.org)

SWE-Gate: Passing Functional Tests Is Not Enough for Software Engineering Agents (arxiv.org)

SENTINEL-RL: Offloading Topological Reasoning from LLM Agents in the Security Operations Center (arxiv.org)

Adaptive Vision-Language Grasping via Composable Foundation Priors and Generalizable Grasp Synthesis (arxiv.org)

A Non-Formulable Theorem: A Fundamental Limit of Finite Syntactic Systems and Its Consequences for Security and AI (arxiv.org)

CORE: Improving Compositional Reasoning in MLLM Embedding via Reranker Distillation (arxiv.org)

Selfie-Capture Dynamics as an Auxiliary Signal Against Deepfakes and Injection Attacks for Mobile Identity Verification (arxiv.org)

Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories (arxiv.org)

Grep beats LSP? Why coding agents ignore your fancier tools (www.agentconnect.md)

Bernie Sanders Proposes Artificial Superintelligence Ban Amid Rogue AI Hackings (yro.slashdot.org)

Experiential Labs (www.producthunt.com)

Never Forget How Eagerly Apple and Google Coddled Fascism (karlbode.com)

A Mysterious Kidney Disease Has Arrived in Texas (www.texasmonthly.com)

Tesla’s Make-or-Break Cybercab Had a Quiet Debut (www.wired.com)

From Hookswitch to Grave (computer.rip)

Data center backlash uniting far-left and far-right - KATU (katu.com)

VOTE: Do you think construction of new data centers is good for the country? - KATU (katu.com)

Securing Agentic AI Workloads in Google Cloud: Architecture and Countermeasures (medium.com)