AI News

AppliedScientist: Automated Scientific Revision Through Iterative AI Reviewing (arxiv.org)

AcquireBound: Runtime Authorization for Resources Acquired by AI Agents (arxiv.org)

AI Persuasion as a Threat to Human Control (arxiv.org)

Another Blueprint In The Wall: How to Ask Frontier AI Like a Kid? (arxiv.org)

Crypto Accounting Bench: Evaluating Frontier and Open-Weight Models on Crypto-Asset Accounting Tasks (arxiv.org)

ANASSA: An Agentic AI Orchestration Framework for Spatial Intelligence (arxiv.org)

One Model, Two Physical Stories: Auditing Misalignment in Multi-Modal World Modeling (arxiv.org)

El Agente Potente: High-Throughput Agentic Atomistic Simulations (arxiv.org)

Self-Orchestrating Language Models: Leveraging Semantic Dependence for Efficient Inference (arxiv.org)

Domain Generalization for Smartphone-Based Human Activity Recognition: A Systematic Analysis of Components and Interactions (arxiv.org)

DiTAR+: Dual Optimization for Robust Autoregressive Diffusion Speech Synthesis (arxiv.org)

Phorecaster365: A Human-Supervised Reference Architecture for Hybrid Pharmaceutical Sales Forecasting and Planning Decision Support (arxiv.org)

When Malicious Instructions Persist: Persistent Memory Poisoning Attack on Harness-Based Agents (arxiv.org)

Assembling the CREW: A Collaborative Multi-agent Reinforcement Learning Framework for Automated Related Work Generation (arxiv.org)

Task-Distribution-Aware Counterweight Synthesis and Constrained Co-Design for Serial Manipulators (arxiv.org)

Salesforce Koa: An Enterprise Language Model for Agentic Tool Use (arxiv.org)

Bridging the Gap in ECG-Based Emotion Recognition: A Unified Evaluation of Deep Learning Models (arxiv.org)

Zero-SNR Analyticity of the Scalar MMSE Is Equivalent to Gaussianity (arxiv.org)

SpliTEE: Improving LLM Inference on Trusted Hardware with Differentially Private GPU Outsourcing (arxiv.org)

Validating Hybrid-State Cache Recovery for GLM-5.3-Flash with vLLM and LMCache (arxiv.org)

ABSOL: Aggregated Bayesian Subsampling Orchestrated with LLMs (arxiv.org)

Learned Bow Control on a Measured Bowed-String Model: a Revised Minimum-Bow-Force Law, a Recurrent Controller, and the Domain of a Supervision Ceiling (arxiv.org)

Linearized PINN with pretrained nonlinear layers (arxiv.org)

GGUF-Metadata Prediction of Single-Sequence llama.cpp Throughput Across Three Systems (arxiv.org)

Externalizing Requirement-to-Repair Artifacts as Observable Traces for LLM-Based Program Repair (arxiv.org)

Geometric Flow enhanced Graph Coarsening (arxiv.org)

Towards a knowledge-enhanced single-cell foundation model (arxiv.org)

MemRiskBench: Trace-Aware Risk-Preserving Evaluation for Long-Horizon LLM Agents (arxiv.org)

Converting Sequenced Fuzzy Cognitive Maps to Causal Virtual Worlds with Large Video Generators (arxiv.org)

Shallow Beliefs: Synthetic document finetuning does not inoculate against emergent misalignment from reward hacking (arxiv.org)

Geometric Signatures of Conceptual Reorganization: A Counterfactual Embedding Framework for Detecting Scientific Revolutions (arxiv.org)

Forty Shades of Blue: Quality-Diversity Alignment via Mode-Conditioned Reinforcement Learning (arxiv.org)

An explicit solution of the five-expert prediction PDE and the exact optimality set of COMB (arxiv.org)

Prescreening Point Defects in Semiconductors With Machine Learning (arxiv.org)

TwinICL: Diagnosing Multimodal In-Context Learning through Paired Counterfactuals (arxiv.org)

Route, Don't Fix: Regime-Dependent Decoding Correction and a Trajectory-Gated Router for Reliable Clinical LLM Answer Selection (arxiv.org)

Mind Which Bird You Favour: Parameterizing Adequacy-Fluency Balance in Meta-Evaluation of Machine Translation (arxiv.org)

A Personalized Dynamic Balance Evaluation Paradigm for Hip Exoskeleton-Assisted Walking under Unexpected Ground Perturbations (arxiv.org)

Bypass Observation: A Conceptual Design of a Non-Intrusive Layer-Wise Semantic Extraction Architecture (arxiv.org)

ViperQ: Order Flow Pattern Recognition via Auction Market Theory for Reinforcement Learning Trading (arxiv.org)

UniCAR-RL: Seeing Better before Thinking Deeper in Visual Mathematics (arxiv.org)

ClinAgent: A ReAct-Based Agent for Conversational Access to Clinical Trial Information (arxiv.org)

CCMAN: Cognitive Instability-Aware Cross-Modal Attention Network for Interpretable Temporal Biomarkers of Verbal Fluency Speech (arxiv.org)

HELENA for 5G NR LEO NTN Channel Estimation: A Comparative Evaluation (arxiv.org)

Parameter isolation with domain-specific experts for incremental audio classification (arxiv.org)

Channel-Adaptive Region Adjacency Graph Carriers for Semantic Image Communication (arxiv.org)

AlgoRAG: Retrieval-Augmented Generation for Theoretical Computer Science Education -- A Comprehensive Evaluation Framework for Algorithm Analysis and Complexity Theory (arxiv.org)

Learning Multi-Agent Task Assignment and Navigation in the Factory: from Simulation to Real Robots (arxiv.org)

Sharing standardized image-derived data in computational pathology using DICOM (arxiv.org)

Theseus in the Graph: Towards Traceable Multi-Hop Graph Navigation (arxiv.org)