arXiv:2608.17163v1 Announce Type: cross Abstract: Off-policy reinforcement learning (RL) has become increasingly sample-efficient, enabling applications…
Category: cs.AI updates on arXiv.org
The Acknowledgment Point Is the System: Durable Policy-Decision Receipts for AI Audit Evidence
arXiv:2608.17176v1 Announce Type: cross Abstract: An AI audit record is useful only if its durability and trust boundary are explicit. Returning a guarded…
Expected free energy as an information constraint on the Bethe Lagrangian
arXiv:2608.17167v1 Announce Type: cross Abstract: Active inference selects actions by minimising an expected free energy functional over predicted…
From Abductive Explanations to Global Logical Rules for Node Classification in SGCs
arXiv:2608.17103v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) have achieved remarkable performance in node classification tasks,…
Foundation Agents Meet Agentic Deep Research: Evidence-Grounded Clinical Code Forecasting
arXiv:2608.17075v1 Announce Type: cross Abstract: Next-encounter ICD forecasting predicts which standardized diagnosis codes will be documented at a…
Iterative tensor network transformations for element-wise evaluation of elementary and filtering functions
arXiv:2608.17135v1 Announce Type: cross Abstract: Tensor networks are powerful formats for compressing large-scale data. However, their application to…
Authorization Before Context: A Model-Neutral Audience Boundary Against Cross-Audience Memory Leakage in Agentic Systems
arXiv:2608.17148v1 Announce Type: cross Abstract: A personal language agent learns a fact from one audience and may later place it in the prompt it…
Structured Driving-State Narratives for Small Language Model-Based GNSS Spoofing Detection
arXiv:2608.17092v1 Announce Type: cross Abstract: Autonomous vehicles (AVs) depend on reliable Global Navigation Satellite System (GNSS) positioning.…
YILDIZ-VPR: A Novel Dataset with Dense Coverage Under Diverse Environmental Conditions for Visual Place Recognition
arXiv:2608.17033v1 Announce Type: cross Abstract: Visual Place Recognition (VPR) aims to recognize the location of a query image by comparing it with a…
Institution-Specific LLM Prompting Recovers PHI That De-identification Systems and Their Gold Standards Both Miss
arXiv:2608.17051v1 Announce Type: cross Abstract: Secondary use of electronic health records requires de-identification, yet existing systems miss…