AI News

Are Classification Robustness and Explanation Robustness Really Strongly Correlated? An Analysis Through Input Loss Landscape (arxiv.org)

Repair Before Veto, When Repair Is Hidden: Quantum-Accessible Features for Repair-Augmented Constraint Learning (arxiv.org)

IEA: Amateur-Friendly Conversational Image Editing Agent via Three Stages of Multitask Alignment (arxiv.org)

GVC-Seg: Training-Free 3D Instance Segmentation via Geometric Visual Correspondence (arxiv.org)

Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation (arxiv.org)

Summarization is Not Dead Yet (arxiv.org)

CATPO: Critique-Augmented Tree Policy Optimization (arxiv.org)

Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook (arxiv.org)

Multi-planar 2D-U-Net Segmentation of 3D-CT Abdominal Organs augmented by Spatial Occurrence Maps (arxiv.org)

Targeting World Models to Compromise Robot Learning Pipelines (arxiv.org)

Context-Aware Deep Learning for Defect Classification in Atomic-Resolution STEM (arxiv.org)

Harness Engineering for Physical AI: Robot Middleware Is the Harness Layer (arxiv.org)

AI Assurance in UK Defence: Challenges in Operationalising JSP 936 (arxiv.org)

Deep reinforcement learning for process design: Review and perspective (arxiv.org)

TAMUNA: Doubly Accelerated Distributed Optimization under Partial Participation (arxiv.org)

Echo-Memory: A Controlled Study of Memory in Action World Models (arxiv.org)

Discovering Functionally Selective Brain Regions with a Deep Topographic Multimodal Model (arxiv.org)

Data Synthesis and Parameter-Efficient Fine-Tuning for Low-Resource NMT: A Case Study on Q'eqchi' Mayan (arxiv.org)

Can Data Work be Reparative? (arxiv.org)

PACT: Learning Diverse Diagnostic Strategies via Privileged Synthesis and Branch Consensus (arxiv.org)

Report on CHIIR 2026 Workshop on Generative AI and Academic Search (GAI&AS) (arxiv.org)

From Statute to Control Flow: Span-Grounded Deontic Trees for Defeasible Scope Parsing (arxiv.org)

SpectraLDS: Provable Distillation for Linear Dynamical Systems (arxiv.org)

Learning from flowsheets: A generative transformer model for autocompletion of flowsheets (arxiv.org)

Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models (arxiv.org)

Hybrid Robustness Verification for Spatio-Temporal Neural Networks (arxiv.org)

Adaptive directional gradients for parameterised quantum circuits (arxiv.org)

PolyBuild: An End-to-End Method for Polygonal Building Contour Extraction from High-Resolution Remote Sensing Images (arxiv.org)

Failure-Aware Refinement of Vision-Language Model for Lithography Defect Detection (arxiv.org)

Few-shot Class-variable Incremental Audio Classification via Prototype Adaptation and Pseudo Class-variable Training (arxiv.org)

A multi-agent system for spine MRI report generation from multi-sequence imaging (arxiv.org)

Benchmarking Vision-Language-Action Models on SO-101: Failure and Recovery Analysis (arxiv.org)

Difference-Aware Retrieval Policies for Imitation Learning (arxiv.org)

What the Eyes See, the LLMs Miss: Exploiting Human Perception for Adversarial Text Attacks (arxiv.org)

End-to-End Context Compression at Scale (arxiv.org)

Do Video Foundation Models Understand Intuitive Physics? A Layerwise Probing Analysis (arxiv.org)

FMplex: Model Virtualization for Serving Extensible Foundation Models (arxiv.org)

Learning to Attack and Defend: Adaptive Red Teaming of Language Models via GRPO (arxiv.org)

ReCoVLA: VLM-Guided Reward Compilation for Failure Recovery in Vision-Language-Action Policies (arxiv.org)

Code Is More Than Text: Uncertainty Estimation for Code Generation (arxiv.org)

Integrating gene regulatory priors into Transformer attention with scTransformer for interpretable scRNA-seq analysis (arxiv.org)

Streaming Interventions: Can Video Large Language Models Correct Mistakes as They Occur? (arxiv.org)

Intelligent Character Recognition of Handwritten Forms with Deep Neural Networks (arxiv.org)

BLM-SGAN: Bidirectional Language Modeling for Semantic-Spatial Text-to-Image Generation (arxiv.org)

Governance Controls for AI-Generated Test Artifacts in Autonomous Software Testing (arxiv.org)

AI-Augmented Closed-Loop Quality Engineering: A Reference Architecture for Continuous Software Quality Intelligence (arxiv.org)

Evaluating AI Investment Strategies (arxiv.org)

See More, Think Deeper: Query-Expanded Visual Evidence and Answer-Clue Guided Reflection for Long Video Understanding (arxiv.org)

Unifying Object-Centric World Models and Diffusion Policy: A Hierarchical Framework for Multi-Stage Robotic Tasks (arxiv.org)

RadOT-Eval: Auditable Structured-Evidence Transport for Radiology Report Evaluation (arxiv.org)