AI News

Token Efficient Task Execution via Application Behavior Modeling for Web Agents (arxiv.org)

ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search (arxiv.org)

Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement (arxiv.org)

Vibe Patenting: Evaluating LLM Judges for Professional Patent-Drafting Agents (arxiv.org)

Toward Self-Adaptive Physical AI: Can LLM Agents Manage Long-Horizon Physical Tasks? (arxiv.org)

Carbon-Aware Routing for Function Calling in Edge-Cloud LLM Systems (arxiv.org)

LabAgent: Customize Any Research Hubs for Scientific Discoveries Using AI Agents (arxiv.org)

TimeThink: Eliciting Compositional Reasoning in Timeseries Large Language Models (arxiv.org)

Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures (arxiv.org)

Governing at Machine Speed: An Adaptive Intelligence Architecture for Real-Time AI Policy Enforcement (arxiv.org)

OrchSLM: Probing the Dynamics of Small Language Model Orchestration (arxiv.org)

Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation (arxiv.org)

From Legal Text to AI-specific Risk Sources: A Systematic Analysis of the EU AI Act's High-Risk Requirements (arxiv.org)

Asclepius: An Adaptive Harness for Long-Horizon Clinical Agents (arxiv.org)

AutoTailor: Automatic, User-Aligned Capability Selection and Adaptation for Web Agents (arxiv.org)

Toward a Decision-Assurance Layer for AI-Assisted Flight Planning in Air Traffic Management (arxiv.org)

A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics (arxiv.org)

Planning or Learning: Reliability and Cost in Multi-Asset Maintenance (arxiv.org)

Causal multi-modal AI for personalized chemosensitivity prediction (arxiv.org)

How User-AI Mistreatment Occurs and Matters in Conversational Systems? (arxiv.org)

FLoKD: Adaptive Knowledge Distillation for Federated Low-Rank LLM over Wireless Networks (arxiv.org)

$\tau$-Elicitation: Benchmarking multi-turn entity extraction in voice agents (arxiv.org)

Identity Is More Than Recall: A Benchmark for Persistent Identity in Deployed AI Agents (arxiv.org)

Solar Intelligence (arxiv.org)

Safety as a Constraint: Fine-Tuning a LLM Recommender to Explain Itself (arxiv.org)

FedV-KGQA in Practice: Design Lessons and an Interactive Prototype (arxiv.org)

Cost Characterization of Vertically Partitioned Federated Knowledge Graphs (arxiv.org)

GeoSkill:Experience-Driven Hierarchical Skill Learning with Collaborative Revision forGeospatialAgents (arxiv.org)

Enhancing Event Candidate Acquisition for Event Linking (arxiv.org)

Recoverability as a System Primitive for Long-Horizon AI Agents (arxiv.org)

Windowed A-K-MDP (arxiv.org)

Drift-Constrained Optimization: Only Direction Matters in Fine-Tuning Instruct Models (arxiv.org)

Degraded but Not Entirely Ineffective: PE-Based Deformable Graph Neural Networks (arxiv.org)

JaxAHT: A JAX-Based Library for Ad Hoc Teamwork (arxiv.org)

MANAS-2: Constrained Reconstruction for EEG Foundation Models (arxiv.org)

Trustworthy Agentic AI: A Comprehensive Cybersecurity and Systems Survey on Threat Landscapes, Defense Architectures, and Open Challenges (arxiv.org)

How Many Thoughts Can a Vector Hold? The Capacity of Reasoning by Superposition (arxiv.org)

Positioning manuscripts in the scientific landscape with agentic AI (arxiv.org)

Homeostatic Continual Learning (arxiv.org)

Surprising Effectiveness of Self-Demonstrations in Enhancing Schema-Ontology Mapping with LLMs (arxiv.org)

Partition Scores Are Not System Scores: Deployment-Fidelity Gaps in Decomposed Algorithm Selection (arxiv.org)

Do Not Restart: Residual Completion for Stateful Agent Handoffs (arxiv.org)

Clean Scores, Buried Evidence, and Confident Wrong: A Receipt-Based Audit of Frontier Agentic QA (arxiv.org)

Operational Range Bounding in Spectroscopy: A Safety Cage Framework for Machine Learning Models (arxiv.org)

Self-Indexing Attention for Compression-Compatible Sparse Long-Context LLM Inference (arxiv.org)

From objective discovery to prediction of global ocean eco-provinces: A pathway for trustworthy learning (arxiv.org)

Part Grounding, Not Action Knowledge: Locating the Bottleneck in VLM Affordance Prediction (arxiv.org)

From Masking to Merging: Rethinking SpecAugment for Efficient Audio Spectrogram Transformer (arxiv.org)

Frame-Synchronous Hand Gesture Detection by Projected Winding Order (arxiv.org)

Speak to the City: Multimodal Resolution for Outside-the-Vehicle References (arxiv.org)