Zirui Zhuang

SAT: Synergistic Adaptive Training for Long-Tailed Recognition

M-LoRA: Efficient Serving for Concurrent LoRA Adapters with Memory-Aware Speculative Scheduler on Single GPU

M-LoRA is proposed, a memory-aware multi-LoRA serving system that reduces queuing delay and improves throughput through efficient request scheduling guided by fine-grained memory modeling.

Do LVLMs Truly Understand Video Anomalies? Revealing Hallucination via Co-Occurrence Patterns

This paper investigates LVLMs’ behavior in VAD from a visual-textual co-occurrence perspective, and proposes VAD-DPO, a direct preference optimization method supervised with counter-example pairs that enhances both anomaly detection and reasoning performance, particularly in scene-dependent scenarios.