Search Results for author: Zhenyu Zhang

Found 67 papers, 32 papers with code

Enhancing Chinese Pre-trained Language Model via Heterogeneous Linguistics Graph

3 code implementations ACL 2022 Yanzeng Li, Jiangxia Cao, Xin Cong, Zhenyu Zhang, Bowen Yu, Hongsong Zhu, Tingwen Liu

Chinese pre-trained language models usually exploit contextual character information to learn representations, while ignoring the linguistics knowledge, e. g., word and sentence information.

Language Modelling Sentence

Pushing Auto-regressive Models for 3D Shape Generation at Capacity and Scalability

no code implementations19 Feb 2024 Xuelin Qian, Yu Wang, Simian Luo, yinda zhang, Ying Tai, Zhenyu Zhang, Chengjie Wang, xiangyang xue, Bo Zhao, Tiejun Huang, Yunsheng Wu, Yanwei Fu

In this paper, we extend auto-regressive models to 3D domains, and seek a stronger ability of 3D shape generation by improving auto-regressive models at capacity and scalability simultaneously.

3D Shape Generation Image Generation +1

Topologies of Reasoning: Demystifying Chains, Trees, and Graphs of Thoughts

no code implementations25 Jan 2024 Maciej Besta, Florim Memedi, Zhenyu Zhang, Robert Gerstenberger, Nils Blach, Piotr Nyczyk, Marcin Copik, Grzegorz Kwaśniewski, Jürgen Müller, Lukas Gianinazzi, Ales Kubicek, Hubert Niewiadomski, Onur Mutlu, Torsten Hoefler

Among these, prompt engineering coupled with structures has emerged as a promising paradigm, with designs such as Chain-of-Thought, Tree of Thoughts, or Graph of Thoughts, in which the overall LLM reasoning is guided by a structure such as a graph.

Mathematical Reasoning Prompt Engineering

QuantumSEA: In-Time Sparse Exploration for Noise Adaptive Quantum Circuits

1 code implementation10 Jan 2024 Tianlong Chen, Zhenyu Zhang, Hanrui Wang, Jiaqi Gu, Zirui Li, David Z. Pan, Frederic T. Chong, Song Han, Zhangyang Wang

To address these two pain points, we propose QuantumSEA, an in-time sparse exploration for noise-adaptive quantum circuits, aiming to achieve two key objectives: (1) implicit circuits capacity during training - by dynamically exploring the circuit's sparse connectivity and sticking a fixed small number of quantum gates throughout the training which satisfies the coherence time and enjoy light noises, enabling feasible executions on real quantum devices; (2) noise robustness - by jointly optimizing the topology and parameters of quantum circuits under real device noise models.

Quantum Machine Learning

Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Intervention

2 code implementations22 Dec 2023 Zhen Tan, Tianlong Chen, Zhenyu Zhang, Huan Liu

Large Language Models (LLMs) have achieved unprecedented breakthroughs in various natural language processing domains.

FFT: Towards Harmlessness Evaluation and Analysis for LLMs with Factuality, Fairness, Toxicity

1 code implementation30 Nov 2023 Shiyao Cui, Zhenyu Zhang, Yilong Chen, Wenyuan Zhang, Tianyun Liu, Siqi Wang, Tingwen Liu

The widespread of generative artificial intelligence has heightened concerns about the potential harms posed by AI-generated texts, primarily stemming from factoid, unfair, and toxic content.

Fairness Instruction Following +1

Simple but Effective Unsupervised Classification for Specified Domain Images: A Case Study on Fungi Images

no code implementations15 Nov 2023 Zhaocong liu, Fa Zhang, Lin Cheng, Huanxi Deng, Xiaoyan Yang, Zhenyu Zhang, ChiChun Zhou

Addressing this, an unsupervised classification method with three key ideas is introduced: 1) dual-step feature dimensionality reduction using a pre-trained model and manifold learning, 2) a voting mechanism from multiple clustering algorithms, and 3) post-hoc instead of prior manual annotation.

Classification Dimensionality Reduction +1

Sam-Guided Enhanced Fine-Grained Encoding with Mixed Semantic Learning for Medical Image Captioning

no code implementations2 Nov 2023 Zhenyu Zhang, Benlu Wang, Weijie Liang, Yizhi Li, Xuechen Guo, Guanhong Wang, Shiyan Li, Gaoang Wang

With the development of multimodality and large language models, the deep learning-based technique for medical image captioning holds the potential to offer valuable diagnostic recommendations.

Image Captioning

Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity

1 code implementation8 Oct 2023 Lu Yin, You Wu, Zhenyu Zhang, Cheng-Yu Hsieh, Yaqing Wang, Yiling Jia, Mykola Pechenizkiy, Yi Liang, Zhangyang Wang, Shiwei Liu

Large Language Models (LLMs), renowned for their remarkable performance across diverse domains, present a challenge when it comes to practical deployment due to their colossal model size.

Network Pruning

Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy

1 code implementation2 Oct 2023 Pingzhi Li, Zhenyu Zhang, Prateek Yadav, Yi-Lin Sung, Yu Cheng, Mohit Bansal, Tianlong Chen

Sparsely activated Mixture-of-Experts (SMoE) has shown promise to scale up the learning capacity of neural networks, however, they have issues like (a) High Memory Usage, due to duplication of the network layers into multiple copies as experts; and (b) Redundancy in Experts, as common learning-based routing policies suffer from representational collapse.

JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention

no code implementations1 Oct 2023 Yuandong Tian, Yiping Wang, Zhenyu Zhang, Beidi Chen, Simon Du

We propose Joint MLP/Attention (JoMA) dynamics, a novel mathematical framework to understand the training procedure of multilayer Transformer architectures.

A study on the impact of pre-trained model on Just-In-Time defect prediction

1 code implementation5 Sep 2023 Yuxiang Guo, Xiaopeng Gao, Zhenyu Zhang, W. K. Chan, Bo Jiang

These findings emphasize the effectiveness of transformer-based pre-trained models in JIT defect prediction tasks, especially in scenarios with limited training data.

Defect Detection

RigNet++: Semantic Assisted Repetitive Image Guided Network for Depth Completion

no code implementations1 Sep 2023 Zhiqiang Yan, Xiang Li, Le Hui, Zhenyu Zhang, Jun Li, Jian Yang

To tackle these challenges, we explore a repetitive design in our image guided network to gradually and sufficiently recover depth values.

Depth Completion Depth Estimation +1

H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models

1 code implementation24 Jun 2023 Zhenyu Zhang, Ying Sheng, Tianyi Zhou, Tianlong Chen, Lianmin Zheng, Ruisi Cai, Zhao Song, Yuandong Tian, Christopher Ré, Clark Barrett, Zhangyang Wang, Beidi Chen

Based on these insights, we propose Heavy Hitter Oracle (H$_2$O), a KV cache eviction policy that dynamically retains a balance of recent and H$_2$ tokens.

Variable Radiance Field for Real-Life Category-Specifc Reconstruction from Single Image

no code implementations8 Jun 2023 Kun Wang, Zhiqiang Yan, Zhenyu Zhang, Xiang Li, Jun Li, Jian Yang

Our key contributions are: (1) We parameterize the geometry and appearance of the object using a multi-scale global feature extractor, which avoids frequent point-wise feature retrieval and camera dependency.

Contrastive Learning Object +1

Are Large Kernels Better Teachers than Transformers for ConvNets?

1 code implementation30 May 2023 Tianjin Huang, Lu Yin, Zhenyu Zhang, Li Shen, Meng Fang, Mykola Pechenizkiy, Zhangyang Wang, Shiwei Liu

We hereby carry out a first-of-its-kind study unveiling that modern large-kernel ConvNets, a compelling competitor to Vision Transformers, are remarkably more effective teachers for small-kernel ConvNets, due to more similar architectures.

Knowledge Distillation

OPDN: Omnidirectional Position-aware Deformable Network for Omnidirectional Image Super-Resolution

no code implementations26 Apr 2023 Xiaopeng Sun, Weiqi Li, Zhenyu Zhang, Qiufang Ma, Xuhan Sheng, Ming Cheng, Haoyu Ma, Shijie Zhao, Jian Zhang, Junlin Li, Li Zhang

Model A aims to enhance the feature extraction ability of 360{\deg} image positional information, while Model B further focuses on the high-frequency information of 360{\deg} images.

Image Super-Resolution Position

Learning Versatile 3D Shape Generation with Improved AR Models

no code implementations26 Mar 2023 Simian Luo, Xuelin Qian, Yanwei Fu, yinda zhang, Ying Tai, Zhenyu Zhang, Chengjie Wang, xiangyang xue

Auto-Regressive (AR) models have achieved impressive results in 2D image generation by modeling joint distributions in the grid space.

3D Shape Generation Image Generation +1

Graph Transformer GANs for Graph-Constrained House Generation

no code implementations CVPR 2023 Hao Tang, Zhenyu Zhang, Humphrey Shi, Bo Li, Ling Shao, Nicu Sebe, Radu Timofte, Luc van Gool

We present a novel graph Transformer generative adversarial network (GTGAN) to learn effective graph node relations in an end-to-end fashion for the challenging graph-constrained house generation task.

Generative Adversarial Network House Generation +1

Sparsity May Cry: Let Us Fail (Current) Sparse Neural Networks Together!

1 code implementation3 Mar 2023 Shiwei Liu, Tianlong Chen, Zhenyu Zhang, Xuxi Chen, Tianjin Huang, Ajay Jaiswal, Zhangyang Wang

In pursuit of a more general evaluation and unveiling the true potential of sparse algorithms, we introduce "Sparsity May Cry" Benchmark (SMC-Bench), a collection of carefully-curated 4 diverse tasks with 10 datasets, that accounts for capturing a wide range of domain-specific and sophisticated knowledge.

Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformers

1 code implementation2 Mar 2023 Tianlong Chen, Zhenyu Zhang, Ajay Jaiswal, Shiwei Liu, Zhangyang Wang

Despite their remarkable achievement, gigantic transformers encounter significant drawbacks, including exorbitant computational and memory footprints during training, as well as severe collapse evidenced by a high degree of parameter redundancy.

Robust Weight Signatures: Gaining Robustness as Easy as Patching Weights?

1 code implementation24 Feb 2023 Ruisi Cai, Zhenyu Zhang, Zhangyang Wang

Given a robust model trained to be resilient to one or multiple types of distribution shifts (e. g., natural image corruptions), how is that "robustness" encoded in the model weights, and how easily can it be disentangled and/or "zero-shot" transferred to some other models?

Learning Neural Proto-Face Field for Disentangled 3D Face Modeling in the Wild

no code implementations CVPR 2023 Zhenyu Zhang, Renwang Chen, Weijian Cao, Ying Tai, Chengjie Wang

To address this problem, this paper presents a novel Neural Proto-face Field (NPF) for unsupervised robust 3D face modeling.

Learning To Measure the Point Cloud Reconstruction Loss in a Representation Space

no code implementations CVPR 2023 Tianxin Huang, Zhonggan Ding, Jiangning Zhang, Ying Tai, Zhenyu Zhang, Mingang Chen, Chengjie Wang, Yong liu

Specifically, we use the contrastive constraint to help CALoss learn a representation space with shape similarity, while we introduce the adversarial strategy to help CALoss mine differences between reconstructed results and ground truths.

Point cloud reconstruction

Towards Generalized Open Information Extraction

no code implementations29 Nov 2022 Bowen Yu, Zhenyu Zhang, Jingyang Li, Haiyang Yu, Tingwen Liu, Jian Sun, Yongbin Li, Bin Wang

Open Information Extraction (OpenIE) facilitates the open-domain discovery of textual facts.

Open Information Extraction

DesNet: Decomposed Scale-Consistent Network for Unsupervised Depth Completion

no code implementations20 Nov 2022 Zhiqiang Yan, Kun Wang, Xiang Li, Zhenyu Zhang, Jun Li, Jian Yang

Unsupervised depth completion aims to recover dense depth from the sparse one without using the ground-truth annotation.

Depth Completion Depth Estimation +2

QuanGCN: Noise-Adaptive Training for Robust Quantum Graph Convolutional Networks

no code implementations9 Nov 2022 Kaixiong Zhou, Zhenyu Zhang, Shengyuan Chen, Tianlong Chen, Xiao Huang, Zhangyang Wang, Xia Hu

Quantum neural networks (QNNs), an interdisciplinary field of quantum computing and machine learning, have attracted tremendous research interests due to the specific quantum advantages.

An Efficient End-to-End Transformer with Progressive Tri-modal Attention for Multi-modal Emotion Recognition

no code implementations20 Sep 2022 Yang Wu, Pai Peng, Zhenyu Zhang, Yanyan Zhao, Bing Qin

At the low-level, we propose the progressive tri-modal attention, which can model the tri-modal feature interactions by adopting a two-pass strategy and can further leverage such interactions to significantly reduce the computation and memory complexity through reducing the input token length.

Emotion Recognition

Layout-Aware Information Extraction for Document-Grounded Dialogue: Dataset, Method and Demonstration

no code implementations14 Jul 2022 Zhenyu Zhang, Bowen Yu, Haiyang Yu, Tingwen Liu, Cheng Fu, Jingyang Li, Chengguang Tang, Jian Sun, Yongbin Li

In this paper, we propose a Layout-aware document-level Information Extraction dataset, LIE, to facilitate the study of extracting both structural and semantic knowledge from visually rich documents (VRDs), so as to generate accurate responses in dialogue systems.

Language Modelling

Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

1 code implementation15 Jun 2022 Tianlong Chen, huan zhang, Zhenyu Zhang, Shiyu Chang, Sijia Liu, Pin-Yu Chen, Zhangyang Wang

Certifiable robustness is a highly desirable property for adopting deep neural networks (DNNs) in safety-critical scenarios, but often demands tedious computations to establish.

Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

1 code implementation9 Jun 2022 Tianlong Chen, Zhenyu Zhang, Sijia Liu, Yang Zhang, Shiyu Chang, Zhangyang Wang

For example, on downstream CIFAR-10/100 datasets, we identify double-win matching subnetworks with the standard, fast adversarial, and adversarial pre-training from ImageNet, at 89. 26%/73. 79%, 89. 26%/79. 03%, and 91. 41%/83. 22% sparsity, respectively.

Transfer Learning

Quarantine: Sparsity Can Uncover the Trojan Attack Trigger for Free

1 code implementation CVPR 2022 Tianlong Chen, Zhenyu Zhang, Yihua Zhang, Shiyu Chang, Sijia Liu, Zhangyang Wang

Trojan attacks threaten deep neural networks (DNNs) by poisoning them to behave normally on most samples, yet to produce manipulated results for inputs attached with a particular trigger.

Network Pruning

Label Anchored Contrastive Learning for Language Understanding

no code implementations NAACL 2022 Zhenyu Zhang, Yuming Zhao, Meng Chen, Xiaodong He

Motivated by this, we propose a novel label anchored contrastive learning approach (denoted as LaCon) for language understanding.

Benchmarking Contrastive Learning +3

Multi-Modal Masked Pre-Training for Monocular Panoramic Depth Completion

no code implementations18 Mar 2022 Zhiqiang Yan, Xiang Li, Kun Wang, Zhenyu Zhang, Jun Li, Jian Yang

To deal with the PDC task, we train a deep network that takes both depth and image as inputs for the dense panoramic depth recovery.

Depth Completion Transfer Learning

The Principle of Diversity: Training Stronger Vision Transformers Calls for Reducing All Levels of Redundancy

1 code implementation CVPR 2022 Tianlong Chen, Zhenyu Zhang, Yu Cheng, Ahmed Awadallah, Zhangyang Wang

However, a "head-to-toe assessment" regarding the extent of redundancy in ViTs, and how much we could gain by thoroughly mitigating such, has been absent for this field.

Sparsity Winning Twice: Better Robust Generalization from More Efficient Training

1 code implementation ICLR 2022 Tianlong Chen, Zhenyu Zhang, Pengjun Wang, Santosh Balachandra, Haoyu Ma, Zehao Wang, Zhangyang Wang

We introduce two alternatives for sparse adversarial training: (i) static sparsity, by leveraging recent results from the lottery ticket hypothesis to identify critical sparse subnetworks arising from the early training; (ii) dynamic sparsity, by allowing the sparse subnetwork to adaptively adjust its connectivity pattern (while sticking to the same sparsity ratio) throughout training.

ASFD: Automatic and Scalable Face Detector

no code implementations26 Jan 2022 Jian Li, Bin Zhang, Yabiao Wang, Ying Tai, Zhenyu Zhang, Chengjie Wang, Jilin Li, Xiaoming Huang, Yili Xia

Along with current multi-scale based detectors, Feature Aggregation and Enhancement (FAE) modules have shown superior performance gains for cutting-edge object detection.

Face Detection object-detection +1

Learning To Restore 3D Face From In-the-Wild Degraded Images

no code implementations CVPR 2022 Zhenyu Zhang, Yanhao Ge, Ying Tai, Xiaoming Huang, Chengjie Wang, Hao Tang, Dongjin Huang, Zhifeng Xie

In-the-wild 3D face modelling is a challenging problem as the predicted facial geometry and texture suffer from a lack of reliable clues or priors, when the input images are degraded.

3D Face Modelling Face Reconstruction

You are caught stealing my winning lottery ticket! Making a lottery ticket claim its ownership

1 code implementation NeurIPS 2021 Xuxi Chen, Tianlong Chen, Zhenyu Zhang, Zhangyang Wang

The lottery ticket hypothesis (LTH) emerges as a promising framework to leverage a special sparse subnetwork (i. e., winning ticket) instead of a full model for both training and inference, that can lower both costs without sacrificing the performance.

FMFCC-A: A Challenging Mandarin Dataset for Synthetic Speech Detection

1 code implementation18 Oct 2021 Zhenyu Zhang, Yewei Gu, Xiaowei Yi, Xianfeng Zhao

As increasing development of text-to-speech (TTS) and voice conversion (VC) technologies, the detection of synthetic speech has been suffered dramatically.

Speech Synthesis Synthetic Speech Detection +1

Improving Distantly-Supervised Named Entity Recognition with Self-Collaborative Denoising Learning

1 code implementation EMNLP 2021 Xinghua Zhang, Bowen Yu, Tingwen Liu, Zhenyu Zhang, Jiawei Sheng, Mengge Xue, Hongbo Xu

Distantly supervised named entity recognition (DS-NER) efficiently reduces labor costs but meanwhile intrinsically suffers from the label noise due to the strong assumption of distant supervision.

Denoising named-entity-recognition +2

MediumVC: Any-to-any voice conversion using synthetic specific-speaker speeches as intermedium features

2 code implementations6 Oct 2021 Yewei Gu, Zhenyu Zhang, Xiaowei Yi, Xianfeng Zhao

To realize any-to-any (A2A) voice conversion (VC), most methods are to perform symmetric self-supervised reconstruction tasks (Xi to Xi), which usually results in inefficient performances due to inadequate feature decoupling, especially for unseen speakers.

Voice Conversion

DialogueBERT: A Self-Supervised Learning based Dialogue Pre-training Encoder

no code implementations22 Sep 2021 Zhenyu Zhang, Tao Guo, Meng Chen

DialogueBERT was pre-trained with 70 million dialogues in real scenario, and then fine-tuned in three different downstream dialogue understanding tasks.

Dialogue Understanding Emotion Recognition +7

RigNet: Repetitive Image Guided Network for Depth Completion

no code implementations29 Jul 2021 Zhiqiang Yan, Kun Wang, Xiang Li, Zhenyu Zhang, Jun Li, Jian Yang

However, blurry guidance in the image and unclear structure in the depth still impede the performance of the image guided frameworks.

Depth Completion Depth Estimation +1

Efficient Lottery Ticket Finding: Less Data is More

1 code implementation6 Jun 2021 Zhenyu Zhang, Xuxi Chen, Tianlong Chen, Zhangyang Wang

We observe that a high-quality winning ticket can be found with training and pruning the dense network on the very compact PrAC set, which can substantially save training iterations for the ticket finding process.

GANs Can Play Lottery Tickets Too

1 code implementation ICLR 2021 Xuxi Chen, Zhenyu Zhang, Yongduo Sui, Tianlong Chen

In this work, we for the first time study the existence of such trainable matching subnetworks in deep GANs.

Image-to-Image Translation

Decentralized Baseband Processing with Gaussian Message Passing Detection for Uplink Massive MU-MIMO Systems

no code implementations22 May 2021 Zhenyu Zhang, Yuanyuan Dong, Keping Long, Xiyuan Wang, Xiaoming Dai

Decentralized baseband processing (DBP) architecture, which partitions the base station antennas into multiple antenna clusters, has been recently proposed to alleviate the excessively high interconnect bandwidth, chip input/output data rates, and detection complexity for massive multi-user multiple-input multiple-output (MU-MIMO) systems.

"BNN - BN = ?": Training Binary Neural Networks without Batch Normalization

1 code implementation16 Apr 2021 Tianlong Chen, Zhenyu Zhang, Xu Ouyang, Zechun Liu, Zhiqiang Shen, Zhangyang Wang

However, the BN layer is costly to calculate and is typically implemented with non-binary parameters, leaving a hurdle for the efficient implementation of BNN training.

Image Classification

Hydrogen-assisted layer-by-layer growth and robust nontrivial topology of stanene films on Bi(111)

no code implementations11 Mar 2021 Liying Zhang, Leiqiang Li, Chenxiao Zhao, Shunfang Li, Jinfeng Jia, Zhenyu Zhang, Yu Jia, Ping Cui

The atomistic growth mechanisms and nontrivial topology of stanene as presented here are also discussed in connection with recent experimental findings.

Materials Science

Robust Overfitting may be mitigated by properly learned smoothening

no code implementations ICLR 2021 Tianlong Chen, Zhenyu Zhang, Sijia Liu, Shiyu Chang, Zhangyang Wang

A recent study (Rice et al., 2020) revealed overfitting to be a dominant phenomenon in adversarially robust training of deep networks, and that appropriate early-stopping of adversarial training (AT) could match the performance gains of most recent algorithmic improvements.

Knowledge Distillation

Document-level Relation Extraction with Dual-tier Heterogeneous Graph

no code implementations COLING 2020 Zhenyu Zhang, Bowen Yu, Xiaobo Shu, Tingwen Liu, Hengzhu Tang, Wang Yubin, Li Guo

Document-level relation extraction (RE) poses new challenges over its sentence-level counterpart since it requires an adequate comprehension of the whole document and the multi-hop reasoning ability across multiple sentences to reach the final result.

Decision Making Document-level Relation Extraction +2

Cannot find the paper you are looking for? You can Submit a new open access paper.