Zobrazeno 1 - 10
of 100
pro vyhledávání: '"Park, Dongmin"'
Mitigating hallucinations of Large Vision Language Models,(LVLMs) is crucial to enhance their reliability for general-purpose assistants. This paper shows that such hallucinations of LVLMs can be significantly exacerbated by preceding user-system dia
Externí odkaz:
http://arxiv.org/abs/2403.10492
Autor:
Park, Dongmin
In this dissertation, we propose a systemic framework that prioritizes informative features and examples to enhance each stage of the development process. Specifically, we prioritize informative features and examples and improve the performance of fe
Externí odkaz:
http://arxiv.org/abs/2403.00013
We propose a novel framework DropTop that suppresses the shortcut bias in online continual learning (OCL) while being adaptive to the varying degree of the shortcut bias incurred by continuously changing environment. By the observed high-attention pr
Externí odkaz:
http://arxiv.org/abs/2312.08677
Autor:
Kim, Doyoung, Yoon, Susik, Park, Dongmin, Lee, Youngjun, Song, Hwanjun, Bang, Jihwan, Lee, Jae-Gil
In real-world continual learning (CL) scenarios, tasks often exhibit intricate and unpredictable semantic shifts, posing challenges for fixed prompt management strategies which are tailored to only handle semantic shifts of uniform degree (i.e., unif
Externí odkaz:
http://arxiv.org/abs/2311.12048
Data pruning, which aims to downsize a large training set into a small informative subset, is crucial for reducing the enormous computational costs of modern deep learning. Though large-scale data collections invariably contain annotation noise and n
Externí odkaz:
http://arxiv.org/abs/2311.01002
Unlabeled data examples awaiting annotations contain open-set noise inevitably. A few active learning studies have attempted to deal with this open-set noise for sample selection by filtering out the noisy examples. However, because focusing on the p
Externí odkaz:
http://arxiv.org/abs/2210.07805
Publikováno v:
AAAI 2022
Online recommender systems should be always aligned with users' current interest to accurately suggest items that each user would like. Since user interest usually evolves over time, the update strategy should be flexible to quickly catch users' curr
Externí odkaz:
http://arxiv.org/abs/2203.10354
Autor:
Park, Dongmin, Jung, Yousung
Publikováno v:
In Chem Catalysis 15 February 2024 4(2)
Real-world data inevitably contains noisy labels, which induce the poor generalization of deep neural networks. It is known that the network typically begins to rapidly memorize false-labeled samples after a certain point of training. Thus, to counte
Externí odkaz:
http://arxiv.org/abs/2012.04337
Deep learning has achieved remarkable success in numerous domains with help from large amounts of big data. However, the quality of data labels is a concern because of the lack of high-quality labels in many real-world scenarios. As noisy labels seve
Externí odkaz:
http://arxiv.org/abs/2007.08199