Yuchen Zhang

CLARA's innovative clip-level approach reveals that fine-grained analysis can dramatically enhance hateful video detection accuracy, outperforming traditional methods.

VOLA achieves a 69.4% mean vulnerability-rank recall, outperforming existing models by effectively translating visual cues into actionable driving attributes for open-world scenarios.

AirFlow achieves up to 11.11% lower forecasting error than existing methods while using a fraction of the parameters and computational resources.

Routing a teacher's knowledge across model components can outperform traditional encoder-based approaches in audio captioning.

SPG-Layout achieves a breakthrough in 3D scene synthesis by generating physically plausible layouts in non-Manhattan environments, outperforming existing methods.

Achieving stable and accurate alignment of generative 3D models with partial observations could redefine the standards for 3D reconstruction in computer vision.

ReMP slashes reconfiguration downtime from minutes to mere seconds, revolutionizing how LLMs adapt to fluctuating workloads.

StepGuard's innovative dual-policy approach not only improves navigation accuracy but also recalibrates single-step errors, setting a new benchmark for web navigation tasks.

IstGPT outperforms traditional anomaly detection methods by leveraging LLMs to model complex dependencies in industrial systems, achieving unprecedented accuracy in real-time threat detection.