What Drives Compositional Generalization in Visual Generative Models?
arXiv:2510.03075v2 Announce Type: replace-cross Abstract: Compositional generalization, the ability to generate novel combinations of known concepts, is a key ingredient…
Psychological Steering in LLMs: An Evaluation of Effectiveness and Trustworthiness
arXiv:2510.04484v1 Announce Type: cross Abstract: The ability to control LLMs’ emulated emotional states and personality traits is essential for enabling…
MedCLM: Learning to Localize and Reason via a CoT-Curriculum in Medical Vision-Language Models
arXiv:2510.04477v1 Announce Type: cross Abstract: Bridging clinical diagnostic reasoning with AI remains a central challenge in medical imaging. We introduce…
MINERVA: Mutual Information Neural Estimation for Supervised Feature Selection
arXiv:2510.02610v2 Announce Type: replace-cross Abstract: Existing feature filters rely on statistical pair-wise dependence metrics to model feature-target relationships, but this…
BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks
arXiv:2510.02418v1 Announce Type: new Abstract: LLM web agents now browse and take actions on the open web, yet current agent…
VarCoNet: A variability-aware self-supervised framework for functional connectome extraction from resting-state fMRI
arXiv:2510.02120v2 Announce Type: replace-cross Abstract: Accounting for inter-individual variability in brain function is key to precision medicine. Here, by considering…
Ergodic Risk Measures: Towards a Risk-Aware Foundation for Continual Reinforcement Learning
arXiv:2510.02945v1 Announce Type: cross Abstract: Continual reinforcement learning (continual RL) seeks to formalize the notions of lifelong learning and endless…
Multimodal Carotid Risk Stratification with Large Vision-Language Models: Benchmarking, Fine-Tuning, and Clinical Insights
arXiv:2510.02922v1 Announce Type: cross Abstract: Reliable risk assessment for carotid atheromatous disease remains a major clinical challenge, as it requires…
SingMOS-Pro: An Comprehensive Benchmark for Singing Quality Assessment
arXiv:2510.01812v2 Announce Type: replace-cross Abstract: Singing voice generation progresses rapidly, yet evaluating singing quality remains a critical challenge. Human subjective…
OR-Toolformer: Modeling and Solving Operations Research Problems with Tool Augmented Large Language Models
arXiv:2510.01253v1 Announce Type: new Abstract: Large language models (LLMs) demonstrate strong mathematical reasoning, but reliance on closed-source APIs for OR…