One Figure, Every Canvas: Editable Flowchart Relayout via Agentic Pipeline Read more: https://arxiv.org/html/2610.06852v1

Your Unlearning Gives You Away: Identifying Erased Concepts in Diffusion Models Read more: https://arxiv.org/html/2610.05601v1

LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets Read more: https://arxiv.org/html/2610.09872v1

SciExam for ENSO: Can AI Agents Build Climate Models? Read more: https://arxiv.org/html/2610.10513v1

SkillPoison: Progressive Skill Poisoning via Successful Experiences Read more: https://arxiv.org/html/2610.07645v1

Emergent Unfaithfulness: How Alignment Training Causes Language Models to Silently Override Task Faithfulness Read more: https://arxiv.org/html/2610.00568v1

IntentCoding: Amplifying User Intent in Code Generation Read more: https://arxiv.org/html/2602.00066v1

Towards Safer Autonomous Driving in an Open World: A Dual-Process Approach Read more: https://arxiv.org/html/2610.04088v1

Recursive Video In-Context Learning for Agentic Robot Read more: https://arxiv.org/html/2610.06843v1

TasteVal: Measuring the Experimental Research Taste of AI Systems Against Human Experts Read more: https://arxiv.org/html/2610.06824v2

Tool-calling retrieval versus vector RAG for a small Greek--English knowledge base: accuracy and robustness to how users type Greek Read more: https://arxiv.org/html/2610.08205v1

On-Premises Multi-Course RAG Tutoring for Business Education: Hardware-Software Trade-offs in a Campus AI Tutor Read more: https://arxiv.org/html/2610.02510v1

Copies or Sources? Measuring How LLM Aggregators Count Restated Evidence in Multi-Agent Systems Read more: https://arxiv.org/html/2610.06192v1

STRIDE: Spatial-Temporal Representation for Interval-conditioned Disease Evolution in Longitudinal Glioblastoma MRI Read more: https://arxiv.org/html/2610.08848v1

EviDent-CBCT: Evidence-Bottlenecked Report Generation from Dental CBCT under Non-Exhaustive Report Supervision Read more: https://arxiv.org/html/2610.02375v1

Don't Judge an LLM Only by Its Activations: Discovering Suppressed Safety Features via Counterfactual Activation Potential Read more: https://arxiv.org/html/2610.05541v1

Efficient Neural Surrogates for Linear Radiation Transport on the Lattice and Hohlraum benchmarks Read more: https://arxiv.org/html/2610.04665v1

Argo-Bench: Evaluating Data Agents on Enterprise-Scale Workflows Read more: https://arxiv.org/html/2610.02122v1

Routing Probes Can Improve Without New Information: An Exact-Null Audit of Uncertainty Beyond Model Outputs Read more: https://arxiv.org/html/2609.38956v1

Transformer-Based Time-Series Inference of Lindblad Dynamics in Open Quantum Systems Read more: https://arxiv.org/html/2610.05647v1

Defense-in-Depth for LLMs: Evaluating Memory Gates Against Activation-Induced and Memory-Induced Sycophancy Read more: https://arxiv.org/html/2610.07403v1

Confidence-Ordering Reversal under Contextual Priors in Neural Decoding Read more: https://arxiv.org/html/2610.08229v1

Adaptive Operator Selection in Bilevel Large Neighborhood Search for Electric Autonomous Dial-a-Ride Problem under Uncertainty Read more: https://arxiv.org/html/2610.04219v1

EgoLAP: Learning from Egocentric Human Data through Language-Action Reasoning Read more: https://arxiv.org/html/2610.08726v1

MedCORE: Criteria-Grounded Clinical Reasoning for Interpretable Medical Image Diagnosis Read more: https://arxiv.org/html/2610.08528v1

A doctrine-grounded visual question answering dataset for Tactical Combat Casualty Care Read more: https://arxiv.org/html/2610.07339v1

Beyond Semantic Similarity: Performance and Costs of Agentic Retrieval for Complex Tasks Read more: https://arxiv.org/html/2610.05750v1

GARDiff: Graph-Aligned Residual Diffusion for Probabilistic Multivariate Time-Series Forecasting Read more: https://arxiv.org/html/2609.37694v1

Learning from Viable Failure Prefixes: Milestone Viability Potential Policy Optimization for Long-Horizon LLM Agents Read more: https://arxiv.org/html/2609.37111v2

Spend Teacher Tokens Where They Matter: Success-Referenced On-Policy Distillation Read more: https://arxiv.org/html/2610.02678v1