11 papers
Submission Responsibility Matters: Role-Aware Submission Quotas under Coauthorship
Furkan Mumcu, Yasin Yilmaz
Author-level submission quotas are increasingly used to control growing peer-review load. Recent coauthorship-sensitive quota rules improve over fixed per-author limits by reducing…
Is Video Anomaly Detection Misframed? Evidence from LLM-Based and Multi-Scene Models
Furkan Mumcu, Michael J. Jones, Anoop Cherian +1
Recent video anomaly detection research has expanded rapidly with an emphasis on general models of normality intended to work across many different scenes. While this focus has led…
Detecting Adversarial Data via Provable Adversarial Noise Amplification
Furkan Mumcu, Yasin Yilmaz
The nonuniform and growing impact of adversarial noise across the layers of deep neural networks has been used in the literature, without a formal mathematical justification, to de…
Robustness of Agentic AI Systems via Adversarially-Aligned Jacobian Regularization
Furkan Mumcu, Yasin Yilmaz
As Large Language Models (LLMs) transition into autonomous multi-agent ecosystems, robust minimax training becomes essential yet remains prone to instability when highly non-linear…
Leveraging Multimodal LLM Descriptions of Activity for Explainable Semi-Supervised Video Anomaly Detection
Furkan Mumcu, Michael J. Jones, Anoop Cherian +1
Existing semi-supervised video anomaly detection (VAD) methods often struggle with detecting complex anomalies involving object interactions and generally lack explainability. To o…
Socially-Weighted Alignment: A Game-Theoretic Framework for Multi-Agent LLM Systems
Furkan Mumcu, Yasin Yilmaz
Deploying large language model (LLM) agents in shared environments introduces a fundamental tension between individual alignment and collective stability: locally rational decision…