1 code implementation • 22 Apr 2024 • Wei Huang, Xudong Ma, Haotong Qin, Xingyu Zheng, Chengtao Lv, Hong Chen, Jie Luo, Xiaojuan Qi, Xianglong Liu, Michele Magno
This exploration holds the potential to unveil new insights and challenges for low-bit quantization of LLaMA3 and other forthcoming LLMs, especially in addressing performance degradation problems that suffer in LLM compression.
1 code implementation • 8 Apr 2024 • Xingyu Zheng, Haotong Qin, Xudong Ma, Mingyuan Zhang, Haojie Hao, Jiakai Wang, Zixiang Zhao, Jinyang Guo, Xianglong Liu
With the advancement of diffusion models (DMs) and the substantially increased computational requirements, quantization emerges as a practical solution to obtain compact and efficient low-bit DMs.
1 code implementation • 1 Mar 2024 • Jinyan Hou, Shan Liu, Ya zhang, Haotong Qin
To tackle these challenges, this paper introduces a novel graph construction method tailored to free-floating traffic mode.
no code implementations • 19 Feb 2024 • Hong Chen, Chengtao Lv, Liang Ding, Haotong Qin, Xiabin Zhou, Yifu Ding, Xuebo Liu, Min Zhang, Jinyang Guo, Xianglong Liu, DaCheng Tao
Large language models (LLMs) have significantly advanced the field of natural language processing, while the expensive memory and computation consumption impede their practical deployment.
1 code implementation • 8 Feb 2024 • Haotong Qin, Xudong Ma, Xingyu Zheng, Xiaoyang Li, Yang Zhang, Shouda Liu, Jie Luo, Xianglong Liu, Michele Magno
This paper proposes a novel IR-QLoRA for pushing quantized LLMs with LoRA to be highly accurate through information retention.
1 code implementation • 6 Feb 2024 • Wei Huang, Yangdong Liu, Haotong Qin, Ying Li, Shiming Zhang, Xianglong Liu, Michele Magno, Xiaojuan Qi
Pretrained large language models (LLMs) exhibit exceptional general language processing capabilities but come with significant demands on memory and computational resources.
no code implementations • 3 Feb 2024 • Zixiang Zhao, Lilun Deng, Haowen Bai, Yukun Cui, Zhipeng Zhang, Yulun Zhang, Haotong Qin, Dongdong Chen, Jiangshe Zhang, Peng Wang, Luc van Gool
Therefore, we introduce a novel fusion paradigm named image Fusion via vIsion-Language Model (FILM), for the first time, utilizing explicit textual information in different source images to guide image fusion.
no code implementations • 14 Dec 2023 • Yi Guo, Yiqian He, Xiaoyang Li, Haotong Qin, Van Tung Pham, Yang Zhang, Shouda Liu
Knowledge Distillation (KD) emerges as one of the most promising compression technologies to run advanced deep neural networks on resource-limited devices.
1 code implementation • 24 Nov 2023 • Zhiteng Li, Yulun Zhang, Jing Lin, Haotong Qin, Jinjin Gu, Xin Yuan, Linghe Kong, Xiaokang Yang
In this work, we propose a Binarized Dual Residual Network (BiDRN), a novel quantization method to estimate the 3D human body, face, and hands parameters efficiently.
no code implementations • 5 Sep 2023 • Wei Huang, Haotong Qin, Yangdong Liu, Jingzhuo Liang, Yulun Zhang, Ying Li, Xianglong Liu
Mixed-precision quantization leverages multiple bit-width architectures to unleash the accuracy and efficiency potential of quantized models.
no code implementations • 4 Aug 2023 • Yisong Xiao, Aishan Liu, Tianyuan Zhang, Haotong Qin, Jinyang Guo, Xianglong Liu
Quantization has emerged as an essential technique for deploying deep neural networks (DNNs) on devices with limited resources.
1 code implementation • 27 Jul 2023 • Haotong Qin, Ge-Peng Ji, Salman Khan, Deng-Ping Fan, Fahad Shahbaz Khan, Luc van Gool
Google's Bard has emerged as a formidable competitor to OpenAI's ChatGPT in the field of conversational AI.
no code implementations • 8 Apr 2023 • Yisong Xiao, Tianyuan Zhang, Shunchang Liu, Haotong Qin
To address this gap, we thoroughly evaluated the robustness of quantized models against various noises (adversarial attacks, natural corruptions, and systematic noises) on ImageNet.
no code implementations • 25 Mar 2023 • Yifu Ding, Haotong Qin, Qinghua Yan, Zhenhua Chai, Junjie Liu, Xiaolin Wei, Xianglong Liu
We find the main reasons lie in (1) the existing calibration metric is inaccurate in measuring the quantization influence for extremely low-bit representation, and (2) the existing quantization paradigm is unfriendly to the power-law distribution of Softmax.
1 code implementation • 26 Jan 2023 • Haotong Qin, Mingyuan Zhang, Yifu Ding, Aoyu Li, Zhongang Cai, Ziwei Liu, Fisher Yu, Xianglong Liu
Network binarization emerges as one of the most promising compression approaches offering extraordinary computation and memory savings by minimizing the bit-width.
1 code implementation • 13 Nov 2022 • Haotong Qin, Xudong Ma, Yifu Ding, Xiaoyang Li, Yang Zhang, Zejun Ma, Jiakai Wang, Jie Luo, Xianglong Liu
We highlight that benefiting from the compact architecture and optimized hardware kernel, BiFSMNv2 can achieve an impressive 25. 1x speedup and 20. 2x storage-saving on edge hardware.
1 code implementation • CVPR 2022 • Jiakai Wang, Zixin Yin, Pengfei Hu, Aishan Liu, Renshuai Tao, Haotong Qin, Xianglong Liu, DaCheng Tao
For the generalization against diverse noises, we inject class-specific identifiable patterns into a confined local patch prior, so that defensive patches could preserve more recognizable features towards specific classes, leading models for better recognition under noises.
1 code implementation • ICLR 2022 • Haotong Qin, Yifu Ding, Mingyuan Zhang, Qinghua Yan, Aishan Liu, Qingqing Dang, Ziwei Liu, Xianglong Liu
The large pre-trained BERT has achieved remarkable performance on Natural Language Processing (NLP) tasks but is also computation and memory expensive.
1 code implementation • 14 Feb 2022 • Haotong Qin, Xudong Ma, Yifu Ding, Xiaoyang Li, Yang Zhang, Yao Tian, Zejun Ma, Jie Luo, Xianglong Liu
Then, to allow the instant and adaptive accuracy-efficiency trade-offs at runtime, we also propose a Thinnable Binarization Architecture to further liberate the acceleration potential of the binarized network from the topology perspective.
no code implementations • 1 Dec 2021 • Haotong Qin
Quantization is emerging as an efficient approach to promote hardware-friendly deep learning and run deep neural networks on resource-limited hardware.
no code implementations • 25 Sep 2021 • Haotong Qin, Xiangguo Zhang, Ruihao Gong, Yifu Ding, Yi Xu, Xianglong Liu
We present a novel Distribution-sensitive Information Retention Network (DIR-Net) that retains the information in the forward and backward propagation by improving internal propagation and introducing external representations.
1 code implementation • 1 Sep 2021 • Haotong Qin, Yifu Ding, Xiangguo Zhang, Jiakai Wang, Xianglong Liu, Jiwen Lu
We first give a theoretical analysis that the diversity of synthetic samples is crucial for the data-free quantization, while in existing approaches, the synthetic data completely constrained by BN statistics experimentally exhibit severe homogenization at distribution and sample levels.
1 code implementation • ICCV 2021 • Renshuai Tao, Yanlu Wei, Xiangjian Jiang, Hainan Li, Haotong Qin, Jiakai Wang, Yuqing Ma, Libo Zhang, Xianglong Liu
In this work, we first present a High-quality X-ray (HiXray) security inspection image dataset, which contains 102, 928 common prohibited items of 8 categories.
2 code implementations • 10 Mar 2021 • Hainan Li, Renshuai Tao, Jun Li, Haotong Qin, Yifu Ding, Shuo Wang, Xianglong Liu
Self-supervised learning is emerged as an efficient method to utilize unlabeled data.
no code implementations • CVPR 2021 • Xiangguo Zhang, Haotong Qin, Yifu Ding, Ruihao Gong, Qinghua Yan, Renshuai Tao, Yuhang Li, Fengwei Yu, Xianglong Liu
Unfortunately, we find that in practice, the synthetic data identically constrained by BN statistics suffers serious homogenization at both distribution level and sample level and further causes a significant performance drop of the quantized model.
1 code implementation • 1 Mar 2021 • Renshuai Tao, Yanlu Wei, Hainan Li, Aishan Liu, Yifu Ding, Haotong Qin, Xianglong Liu
The images are gathered from an airport and these prohibited items are annotated manually by professional inspectors, which can be used as a benchmark for model training and further facilitate future research.
1 code implementation • 3 Dec 2020 • Aishan Liu, Shiyu Tang, Xinyun Chen, Lei Huang, Haotong Qin, Xianglong Liu, DaCheng Tao
In this paper, we observe that different $\ell_p$ bounded adversarial perturbations induce different statistical properties that can be separated and characterized by the statistics of Batch Normalization (BN).
1 code implementation • ICLR 2021 • Haotong Qin, Zhongang Cai, Mingyuan Zhang, Yifu Ding, Haiyu Zhao, Shuai Yi, Xianglong Liu, Hao Su
To alleviate the resource constraint for real-time point cloud applications that run on edge devices, in this paper we present BiPointNet, the first model binarization approach for efficient deep learning on point clouds.
2 code implementations • 31 Mar 2020 • Haotong Qin, Ruihao Gong, Xianglong Liu, Xiao Bai, Jingkuan Song, Nicu Sebe
The binary neural network, largely saving the storage and computation, serves as a promising technique for deploying deep models on resource-limited devices.
2 code implementations • CVPR 2020 • Haotong Qin, Ruihao Gong, Xianglong Liu, Mingzhu Shen, Ziran Wei, Fengwei Yu, Jingkuan Song
Our empirical study indicates that the quantization brings information loss in both forward and backward propagation, which is the bottleneck of training accurate binary neural networks.