Zobrazeno 1 - 10
of 54
pro vyhledávání: '"Basu, Sugato"'
Autor:
Li, Jiachen, Feng, Weixi, Fu, Tsu-Jui, Wang, Xinyi, Basu, Sugato, Chen, Wenhu, Wang, William Yang
Diffusion-based text-to-video (T2V) models have achieved significant success but continue to be hampered by the slow sampling speed of their iterative sampling processes. To address the challenge, consistency models have been proposed to facilitate f
Externí odkaz:
http://arxiv.org/abs/2405.18750
Autor:
Jia, Zhiwei, Narayana, Pradyumna, Akula, Arjun R., Pruthi, Garima, Su, Hao, Basu, Sugato, Jampani, Varun
Image ad understanding is a crucial task with wide real-world applications. Although highly challenging with the involvement of diverse atypical scenes, real-world entities, and reasoning over scene-texts, how to interpret image ads is relatively und
Externí odkaz:
http://arxiv.org/abs/2305.18373
Autor:
Feng, Weixi, Zhu, Wanrong, Fu, Tsu-jui, Jampani, Varun, Akula, Arjun, He, Xuehai, Basu, Sugato, Wang, Xin Eric, Wang, William Yang
Attaining a high degree of user controllability in visual generation often requires intricate, fine-grained inputs like layouts. However, such inputs impose a substantial burden on users when compared to simple text inputs. To address the issue, we s
Externí odkaz:
http://arxiv.org/abs/2305.15393
Autor:
He, Xuehai, Feng, Weixi, Fu, Tsu-Jui, Jampani, Varun, Akula, Arjun, Narayana, Pradyumna, Basu, Sugato, Wang, William Yang, Wang, Xin Eric
Diffusion models, such as Stable Diffusion, have shown incredible performance on text-to-image generation. Since text-to-image generation often requires models to generate visual concepts with fine-grained details and attributes specified in text pro
Externí odkaz:
http://arxiv.org/abs/2305.10722
Autor:
Akula, Arjun R., Driscoll, Brendan, Narayana, Pradyumna, Changpinyo, Soravit, Jia, Zhiwei, Damle, Suyash, Pruthi, Garima, Basu, Sugato, Guibas, Leonidas, Freeman, William T., Li, Yuanzhen, Jampani, Varun
Creativity is an indispensable part of human cognition and also an inherent part of how we make sense of the world. Metaphorical abstraction is fundamental in communicating creative ideas through nuanced relationships between abstract concepts such a
Externí odkaz:
http://arxiv.org/abs/2212.09898
Autor:
Feng, Weixi, He, Xuehai, Fu, Tsu-Jui, Jampani, Varun, Akula, Arjun, Narayana, Pradyumna, Basu, Sugato, Wang, Xin Eric, Wang, William Yang
Large-scale diffusion models have achieved state-of-the-art results on text-to-image synthesis (T2I) tasks. Despite their ability to generate high-quality yet creative images, we observe that attribution-binding and compositional capabilities are sti
Externí odkaz:
http://arxiv.org/abs/2212.05032
Autor:
He, Xuehai, Yang, Diji, Feng, Weixi, Fu, Tsu-Jui, Akula, Arjun, Jampani, Varun, Narayana, Pradyumna, Basu, Sugato, Wang, William Yang, Wang, Xin Eric
Prompt tuning is a new few-shot transfer learning technique that only tunes the learnable prompt for pre-trained vision and language models such as CLIP. However, existing prompt tuning methods tend to learn spurious or entangled representations, whi
Externí odkaz:
http://arxiv.org/abs/2210.10362
Autor:
Zhu, Wanrong, Qi, Yuankai, Narayana, Pradyumna, Sone, Kazoo, Basu, Sugato, Wang, Xin Eric, Wu, Qi, Eckstein, Miguel, Wang, William Yang
Vision-and-language navigation (VLN) is a multimodal task where an agent follows natural language instructions and navigates in visual environments. Multiple setups have been proposed, and researchers apply new model architectures or training techniq
Externí odkaz:
http://arxiv.org/abs/2103.16561
The area of constrained clustering has been extensively explored by researchers and used by practitioners. Constrained clustering formulations exist for popular algorithms such as k-means, mixture models, and spectral clustering but have several limi
Externí odkaz:
http://arxiv.org/abs/2101.02792
Autor:
Zhu, Wanrong, Wang, Xin Eric, Narayana, Pradyumna, Sone, Kazoo, Basu, Sugato, Wang, William Yang
A major challenge in visually grounded language generation is to build robust benchmark datasets and models that can generalize well in real-world settings. To do this, it is critical to ensure that our evaluation protocols are correct, and benchmark
Externí odkaz:
http://arxiv.org/abs/2010.03644