AI News

Can Linguistic Reasoning Vectors Enhance Multimodal Reasoning Ability? (arxiv.org)

A Unified Account of Concepts and Chunks (arxiv.org)

Toward AI-Augmented Cooperative Engineering Workflows: Requirements and Architecture the European Rover Challenge (arxiv.org)

AtomWorld-Mem: Memory-Restored World States for Long-Horizon Atomistic Evolution (arxiv.org)

Monitor Jailbreaking: Evading Chain-of-Thought Monitoring Without Encoded Reasoning (arxiv.org)

OmouAI: Argumentative Human-AI Policy Deliberation with Simulated Personas (arxiv.org)

Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents (arxiv.org)

Externalized CPDAG Summaries Improve LLM Causal Deduction (arxiv.org)

HARDEN: Constrained Evolutionary Search for Harder, Answer-Preserving Evaluation Cases (arxiv.org)

Same Text, Different Numbers: The Divergence of LLM-Based Measures (arxiv.org)

Factorized axis convolutional gated recurrent unit with dynamic adaptive pooling for remaining useful life prediction of rolling bearings (arxiv.org)

SciHorizon-eLab: An Agentic Protocol-to-Task Compiler for Scalable Benchmarking of Scientific Embodied Agents (arxiv.org)

Coding Agents Aren't Enough! Evaluating an Enterprise Security Brain for Agentic Cloud Investigations (arxiv.org)

MoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy, Safety, and Tokens (arxiv.org)

FTB Graph: Determining and Validating First-token Broadcasters and Language-Identity Head Circuits in Multilingual Language Models (arxiv.org)

LogicTree-RAG: Logic Tree-guided Retrieval-Augmented Generation for Long-form Patent Drafting (arxiv.org)

Financial Fragility in Societies of LLM Agents: Coordination Failures and Stabilizing Mechanisms (arxiv.org)

MACBT: A Multi-Agent Cognitive Behavioral Therapy Decision Support System with Longitudinal Memory (arxiv.org)

Game Arena: Strategic LLM Evaluation in Competitive Environments (arxiv.org)

Self-Play Search Distillation for Large Language Model Reasoning (arxiv.org)

Training Graph Foundation Models on The Web Graph (arxiv.org)

From Tapping to Hopping: Augmenting Mobile GUI Agents with App-Native Deeplinks (arxiv.org)

EXAONE Demand 1.0: A Time Series Foundation Model for Demand Forecasting (arxiv.org)

TISD: On-Policy Self-Distillation with Trajectory Intervention (arxiv.org)

SkillEvoReg: Regularizing Agent Skill Evolution Against Overfitting (arxiv.org)

Why Jailbreaks Succeed in Diffusion Language Models: An Energy Landscape Analysis (arxiv.org)

PTC-Decoder: Towards Intelligent SLMs on Offline Resource-Constrained Edge Devices (arxiv.org)

A Benchmark and Diagnostic Study of Epistemic Admission in Shared Agent Memory (arxiv.org)

BAT-CLIP: Trimodal Alignment of Brain, Audio and Text (arxiv.org)

Evaluating Real-Time Voice Agents: From Component Quality to Grounded Outcomes (arxiv.org)

HasMem: Hard-Origin Adaptively Softened Memory for Long-Term LLM Agents (arxiv.org)

ConsultMind:Towards Automated Diagnostic Consultation via Uncertainty-Aware Reasoning (arxiv.org)

Does Thinking Help Fairness? Reasoning Tokens Resolve Some Biases but Create More (arxiv.org)

Insurance Reserve Intelligence Platform (arxiv.org)

Neuralyzing the Trace: Selective Representation-Level Unlearning with Contrastive Sparse Autoencoders (arxiv.org)

DanLing NestedTensor: Composable Multi-Ragged Tensors for Deep Learning (arxiv.org)

Benchy: towards a universal language for task-oriented AI benchmarks (arxiv.org)

BioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering (arxiv.org)

Do LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework (arxiv.org)

Pretrained ASR Pseudo-labeling for Noisy Police Audio (arxiv.org)

HCOE: Hyperbolic Clinical Ontology Embeddings from Biomedical Language Models (arxiv.org)

Selective Amortization of Full-Budget Counterfactual Reasoning for Visual Token Communication (arxiv.org)

ORCA: Evaluating LLMs on Data Science Code Translation (arxiv.org)

From S3Q Theory to Implementation: Towards an Architecture for Machine Qualia (arxiv.org)

Learning What to Skip: Counterfactual Credit Assignment for Efficient Multi-Agent LLM Workflows (arxiv.org)

Cheap, open agents make LLM pollution harder to mitigate (arxiv.org)

Analyzing and Mitigating Cost-Inefficient Behaviors in Coding Agents (arxiv.org)

CRC-Router: Risk-Constrained Routing for Medical Agentic AI Systems (arxiv.org)

LAVOIR: Teaching a Single-Pass Decision Encoder When and What to Ask with Amortized Value of Information (arxiv.org)

DeepEdu-v1: Efficient and Scalable Agentic LLMs for Vietnamese Education (arxiv.org)