6 papers
AVENUE: Audio-Video EditiNg Understanding and Evaluation
Hayeon Kim, Yoojin Jang, Jaejun Yoo
Audio-video (AV) editing aims to modify audio and video content according to a target prompt. Unlike single-modality editing, AV editing requires models to infer a modality-selecti…
B-RIGHT: Benchmark Re-evaluation for Integrity in Generalized Human-Object Interaction Testing
Yoojin Jang, Junsu Kim, Hayeon Kim +4
Human-object interaction (HOI) is an essential problem in artificial intelligence (AI) which aims to understand the visual world that involves complex relationships between humans…
Nickel and Diming Your GAN: A Dual-Method Approach to Enhancing GAN Efficiency via Knowledge Distillation
Sangyeop Yeo, Yoojin Jang, Jaejun Yoo
In this paper, we address the challenge of compressing generative adversarial networks (GANs) for deployment in resource-constrained environments by proposing two novel methodologi…
TopP&R: Robust Support Estimation Approach for Evaluating Fidelity and Diversity in Generative Models
Pum Jun Kim, Yoojin Jang, Jisu Kim +1
We propose a robust and reliable evaluation metric for generative models by introducing topological and statistical treatments for rigorous support estimation. Existing metrics, su…
Can We Find Strong Lottery Tickets in Generative Models?
Sangyeop Yeo, Yoojin Jang, Jy-yong Sohn +2
Yes. In this paper, we investigate strong lottery tickets in generative models, the subnetworks that achieve good generative performance without any weight update. Neural network p…
Best-response dynamics, playing sequences, and convergence to equilibrium in random games
Torsten Heinrich, Yoojin Jang, Luca Mungo +4
We analyze the performance of the best-response dynamic across all normal-form games using a random games approach. The playing sequence -- the order in which players update their…