Zobrazeno 1 - 10
of 12
pro vyhledávání: '"Materzynska, Joanna"'
We introduce AirLetters, a new video dataset consisting of real-world videos of human-generated, articulated motions. Specifically, our dataset requires a vision model to predict letters that humans draw in the air. Unlike existing video datasets, ac
Externí odkaz:
http://arxiv.org/abs/2410.02921
Autor:
Longpre, Shayne, Mahari, Robert, Lee, Ariel, Lund, Campbell, Oderinwale, Hamidah, Brannon, William, Saxena, Nayan, Obeng-Marnu, Naana, South, Tobin, Hunter, Cole, Klyman, Kevin, Klamm, Christopher, Schoelkopf, Hailey, Singh, Nikhil, Cherep, Manuel, Anis, Ahmad, Dinh, An, Chitongo, Caroline, Yin, Da, Sileo, Damien, Mataciunas, Deividas, Misra, Diganta, Alghamdi, Emad, Shippole, Enrico, Zhang, Jianguo, Materzynska, Joanna, Qian, Kun, Tiwary, Kush, Miranda, Lester, Dey, Manan, Liang, Minnie, Hamdy, Mohammed, Muennighoff, Niklas, Ye, Seonghyeon, Kim, Seungone, Mohanty, Shrestha, Gupta, Vipul, Sharma, Vivek, Chien, Vu Minh, Zhou, Xuhui, Li, Yizhi, Xiong, Caiming, Villa, Luis, Biderman, Stella, Li, Hanlin, Ippolito, Daphne, Hooker, Sara, Kabbara, Jad, Pentland, Sandy
General-purpose artificial intelligence (AI) systems are built on massive swathes of public web data, assembled into corpora such as C4, RefinedWeb, and Dolma. To our knowledge, we conduct the first, large-scale, longitudinal audit of the consent pro
Externí odkaz:
http://arxiv.org/abs/2407.14933
Autor:
Materzynska, Joanna, Sivic, Josef, Shechtman, Eli, Torralba, Antonio, Zhang, Richard, Russell, Bryan
We introduce an approach for augmenting text-to-video generation models with customized motions, extending their capabilities beyond the motions depicted in the original training data. By leveraging a few video samples demonstrating specific movement
Externí odkaz:
http://arxiv.org/abs/2312.04966
We present a method to create interpretable concept sliders that enable precise control over attributes in image generations from diffusion models. Our approach identifies a low-rank parameter direction corresponding to one concept while minimizing i
Externí odkaz:
http://arxiv.org/abs/2311.12092
Autor:
Schwettmann, Sarah, Shaham, Tamar Rott, Materzynska, Joanna, Chowdhury, Neil, Li, Shuang, Andreas, Jacob, Bau, David, Torralba, Antonio
Publikováno v:
NeurIPS 2023
Labeling neural network submodules with human-legible descriptions is useful for many downstream tasks: such descriptions can surface failures, guide interventions, and perhaps even explain important model behaviors. To date, most mechanistic descrip
Externí odkaz:
http://arxiv.org/abs/2309.03886
Text-to-image models suffer from various safety issues that may limit their suitability for deployment. Previous methods have separately addressed individual issues of bias, copyright, and offensive content in text-to-image models. However, in the re
Externí odkaz:
http://arxiv.org/abs/2308.14761
Motivated by recent advancements in text-to-image diffusion, we study erasure of specific concepts from the model's weights. While Stable Diffusion has shown promise in producing explicit or realistic artwork, it has raised concerns regarding its pot
Externí odkaz:
http://arxiv.org/abs/2303.07345
The CLIP network measures the similarity between natural text and images; in this work, we investigate the entanglement of the representation of word images and natural images in its image encoder. First, we find that the image encoder has an ability
Externí odkaz:
http://arxiv.org/abs/2206.07835
Human action is naturally compositional: humans can easily recognize and perform actions with objects that are different from those used in training demonstrations. In this paper, we study the compositionality of action by looking into the dynamics o
Externí odkaz:
http://arxiv.org/abs/1912.09930
Autor:
Goyal, Raghav, Kahou, Samira Ebrahimi, Michalski, Vincent, Materzyńska, Joanna, Westphal, Susanne, Kim, Heuna, Haenel, Valentin, Fruend, Ingo, Yianilos, Peter, Mueller-Freitag, Moritz, Hoppe, Florian, Thurau, Christian, Bax, Ingo, Memisevic, Roland
Neural networks trained on datasets such as ImageNet have led to major advances in visual object classification. One obstacle that prevents networks from reasoning more deeply about complex scenes and situations, and from integrating visual knowledge
Externí odkaz:
http://arxiv.org/abs/1706.04261