1 code implementation • 30 Nov 2024 • Bytedance-Seed-Foundation-Code-Team, :, Yao Cheng, Jianfeng Chen, Jie Chen, Li Chen, Liyu Chen, Wentao Chen, Zhengyu Chen, Shijie Geng, Aoyan Li, Bo Li, Bowen Li, Linyi Li, Boyi Liu, Jerry Liu, Kaibo Liu, Qi Liu, Shukai Liu, Siyao Liu, Tianyi Liu, Tingkai Liu, Yongfei Liu, Rui Long, Jing Mai, Guanghan Ning, Z. Y. Peng, Kai Shen, Jiahao Su, Jing Su, Tao Sun, Yifan Sun, Yunzhe Tao, Guoyin Wang, Siwei Wang, Xuwu Wang, Yite Wang, Zihan Wang, Jinxiang Xia, Liang Xiang, Xia Xiao, Yongsheng Xiao, Chenguang Xi, Shulin Xin, Jingjing Xu, Shikun Xu, Hongxia Yang, Jack Yang, Yingxiang Yang, Jianbo Yuan, Jun Zhang, Yufeng Zhang, Yuyu Zhang, Shen Zheng, He Zhu, Ming Zhu
As the capabilities of code large language models (LLMs) continue to expand, their applications across diverse code intelligence domains are rapidly increasing.
no code implementations • 18 Aug 2024 • Jiahao Su, Kang You, Zekai Xu, Weizhi Xu, Zhezhi He
When SNN encounters sequence learning, the situation becomes worse due to the difficulties in modeling long-range dependencies.
no code implementations • 7 Jan 2024 • Tahseen Rabbani, Jiahao Su, Xiaoyu Liu, David Chan, Geoffrey Sangston, Furong Huang
Modern ConvNets continue to achieve state-of-the-art results over a vast array of vision and image classification tasks, but at the cost of increasing parameters.
no code implementations • 20 Dec 2023 • Guodong Du, HaoJian Deng, Jiahao Su, Yuan Huang
To be specific, we generate rainy RAW data by converting color rain streak into RAW space and design simple but efficient RAW processing algorithms to synthesize both rainy and clean color images.
1 code implementation • 12 Oct 2023 • Yite Wang, Jiahao Su, Hanlin Lu, Cong Xie, Tianyi Liu, Jianbo Yuan, Haibin Lin, Ruoyu Sun, Hongxia Yang
Our empirical results demonstrate that LEMON reduces computational costs by 56. 7% for Vision Transformers and 33. 2% for BERT when compared to training from scratch.
no code implementations • 13 Jun 2023 • Peijian Ding, Davit Soselia, Thomas Armstrong, Jiahao Su, Furong Huang
While the self-attention operator in vision transformers (ViT) is permutation-equivariant and thus shift-equivariant, patch embedding, positional encoding, and subsampled attention in ViT variants can disrupt this property, resulting in inconsistent predictions even under small shift perturbations.
1 code implementation • 2 Nov 2022 • Kaiwen Yang, Yanchao Sun, Jiahao Su, Fengxiang He, Xinmei Tian, Furong Huang, Tianyi Zhou, DaCheng Tao
In experiments, we show that our method consistently brings non-trivial improvements to the three aforementioned learning tasks from both efficiency and final performance, either or not combined with strong pre-defined augmentations, e. g., on medical images when domain knowledge is unavailable and the existing augmentation techniques perform poorly.
no code implementations • 29 Sep 2021 • Jiahao Su, Wonmin Byeon, Furong Huang
Some of these designs are not exactly orthogonal, while others only consider standard convolutional layers and propose specific classes of their realizations.
no code implementations • ICLR 2022 • Xiaoyu Liu, Jiahao Su, Furong Huang
Guided by tensor diagram representations, we formulate a design space where we can analyze the expressive power of the network structure, providing new directions and possibilities for enhanced performance.
no code implementations • 1 Aug 2021 • Huimin Zeng, Jiahao Su, Furong Huang
Randomized Smoothing (RS), being one of few provable defenses, has been showing great effectiveness and scalability in terms of defending against $\ell_2$-norm adversarial perturbations.
no code implementations • 16 Jun 2021 • Jiahao Su, Wonmin Byeon, Furong Huang
To address this problem, we propose a theoretical framework for orthogonal convolutional layers, which establishes the equivalence between various orthogonal convolutional layers in the spatial domain and the paraunitary systems in the spectral domain.
2 code implementations • NeurIPS 2020 • Jiahao Su, Wonmin Byeon, Jean Kossaifi, Furong Huang, Jan Kautz, Animashree Anandkumar
Learning from spatio-temporal data has numerous applications such as human-behavior analysis, object tracking, video compression, and physics simulation. However, existing methods still perform poorly on challenging video tasks such as long-term forecasting.
Ranked #1 on
Video Prediction
on KTH
(Cond metric)
1 code implementation • NeurIPS 2020 • Jiahao Su, Shiqi Wang, Furong Huang
In this work, we propose to replace any traditional convolutional layer with an autoregressive moving-average (ARMA) layer, a novel module with an adjustable receptive field controlled by the learnable autoregressive coefficients.
no code implementations • 14 Jan 2020 • Jingling Li, Yanchao Sun, Jiahao Su, Taiji Suzuki, Furong Huang
Recently proposed complexity measures have provided insights to understanding the generalizability in neural networks from perspectives of PAC-Bayes, robustness, overparametrization, compression and so on.
no code implementations • ICLR 2020 • Jiahao Su, Milan Cvitkovic, Furong Huang
Bayesian learning of model parameters in neural networks is important in scenarios where estimates with well-calibrated uncertainty are important.
no code implementations • 25 Sep 2019 • Jiahao Su, Wonmin Byeon, Furong Huang, Jan Kautz, Animashree Anandkumar
Long-term video prediction is highly challenging since it entails simultaneously capturing spatial and temporal information across a long range of image frames. Standard recurrent models are ineffective since they are prone to error propagation and cannot effectively capture higher-order correlations.
no code implementations • 25 May 2018 • Jiahao Su, Jingling Li, Bobby Bhattacharjee, Furong Huang
We propose tensorial neural networks (TNNs), a generalization of existing neural networks by extending tensor operations on low order operands to those on high order ones.