Search Results for author: Xiangxiang Chu

Found 46 papers, 29 papers with code

LogicalDefender: Discovering, Extracting, and Utilizing Common-Sense Knowledge

no code implementations18 Mar 2024 Yuhe Liu, Mengxue Kang, Zengchang Qin, Xiangxiang Chu

Experiments show that our model has achieved better logical performance, and the extracted logical knowledge can be effectively applied to other scenarios.

Common Sense Reasoning

PeLK: Parameter-efficient Large Kernel ConvNets with Peripheral Convolution

no code implementations12 Mar 2024 Honghao Chen, Xiangxiang Chu, Yongjian Ren, Xin Zhao, Kaiqi Huang

Due to these issues, current CNNs compromise to scale up to 51x51 in the form of stripe convolution (i. e., 51x5 + 5x51) and start to saturate as the kernel size continues growing.

object-detection Object Detection +1

VisionLLaMA: A Unified LLaMA Interface for Vision Tasks

1 code implementation1 Mar 2024 Xiangxiang Chu, Jianlin Su, Bo Zhang, Chunhua Shen

Large language models are built on top of a transformer-based architecture to process textual inputs.

Image Classification Image Generation +2

MobileVLM V2: Faster and Stronger Baseline for Vision Language Model

1 code implementation6 Feb 2024 Xiangxiang Chu, Limeng Qiao, Xinyu Zhang, Shuang Xu, Fei Wei, Yang Yang, Xiaofei Sun, Yiming Hu, Xinyang Lin, Bo Zhang, Chunhua Shen

We introduce MobileVLM V2, a family of significantly improved vision language models upon MobileVLM, which proves that a delicate orchestration of novel architectural design, an improved training scheme tailored for mobile VLMs, and rich high-quality dataset curation can substantially benefit VLMs' performance.

AutoML Language Modelling

LiDAR-PTQ: Post-Training Quantization for Point Cloud 3D Object Detection

1 code implementation29 Jan 2024 Sifan Zhou, Liang Li, Xinyu Zhang, Bo Zhang, Shipeng Bai, Miao Sun, Ziyu Zhao, Xiaobo Lu, Xiangxiang Chu

To our knowledge, for the very first time in lidar-based 3D detection tasks, the PTQ INT8 model's accuracy is almost the same as the FP32 model while enjoying $3\times$ inference speedup.

3D Object Detection Autonomous Vehicles +3

SCTNet: Single-Branch CNN with Transformer Semantic Information for Real-Time Segmentation

1 code implementation28 Dec 2023 Zhengze Xu, Dongyue Wu, Changqian Yu, Xiangxiang Chu, Nong Sang, Changxin Gao

Recent real-time semantic segmentation methods usually adopt an additional semantic branch to pursue rich long-range context.

Real-Time Semantic Segmentation

Lenna: Language Enhanced Reasoning Detection Assistant

1 code implementation5 Dec 2023 Fei Wei, Xinyu Zhang, Ailing Zhang, Bo Zhang, Xiangxiang Chu

To evaluate the reasoning capability of Lenna, we construct a ReasonDet dataset to measure its performance on reasoning-based detection.

World Knowledge

RobustCalib: Robust Lidar-Camera Extrinsic Calibration with Consistency Learning

no code implementations2 Dec 2023 Shuang Xu, Sifan Zhou, Zhi Tian, Jizhou Ma, Qiong Nie, Xiangxiang Chu

Current traditional methods for LiDAR-camera extrinsics estimation depend on offline targets and human efforts, while learning-based approaches resort to iterative refinement for calibration results, posing constraints on their generalization and application in on-board systems.

Masked Autoencoders Are Robust Neural Architecture Search Learners

no code implementations20 Nov 2023 Yiming Hu, Xiangxiang Chu, Bo Zhang

Neural Architecture Search (NAS) currently relies heavily on labeled data, which is both expensive and time-consuming to acquire.

Image Reconstruction Neural Architecture Search

Norm Tweaking: High-performance Low-bit Quantization of Large Language Models

no code implementations6 Sep 2023 Liang Li, Qingyuan Li, Bo Zhang, Xiangxiang Chu

On GLM-130B and OPT-66B, our method even achieves the same level of accuracy at 2-bit quantization as their float ones.

Model Compression Quantization

FPTQ: Fine-grained Post-Training Quantization for Large Language Models

no code implementations30 Aug 2023 Qingyuan Li, Yifan Zhang, Liang Li, Peng Yao, Bo Zhang, Xiangxiang Chu, Yerui Sun, Li Du, Yuchen Xie

In this study, we propose a novel W4A8 post-training quantization method for the available open-sourced LLMs, which combines the advantages of both two recipes.

Quantization

EfficientRep:An Efficient Repvgg-style ConvNets with Hardware-aware Neural Network Design

1 code implementation1 Feb 2023 Kaiheng Weng, Xiangxiang Chu, Xiaoming Xu, Junshi Huang, Xiaoming Wei

Thus, how to design a neural network to efficiently use the computing ability and memory bandwidth of hardware is a critical problem.

object-detection Object Detection

YOLOv6 v3.0: A Full-Scale Reloading

5 code implementations13 Jan 2023 Chuyi Li, Lulu Li, Yifei Geng, Hongliang Jiang, Meng Cheng, Bo Zhang, Zaidan Ke, Xiaoming Xu, Xiangxiang Chu

For a glimpse of performance, our YOLOv6-N hits 37. 5% AP on the COCO dataset at a throughput of 1187 FPS tested with an NVIDIA Tesla T4 GPU.

Real-Time Object Detection

Make RepVGG Greater Again: A Quantization-aware Approach

2 code implementations3 Dec 2022 Xiangxiang Chu, Liang Li, Bo Zhang

Nonetheless, its quantization performance is usually too poor to deploy (more than 20% top-1 accuracy drop on ImageNet) when INT8 inference is desired.

Quantization Semantic Segmentation

AeDet: Azimuth-invariant Multi-view 3D Object Detection

1 code implementation CVPR 2023 Chengjian Feng, Zequn Jie, Yujie Zhong, Xiangxiang Chu, Lin Ma

However, the typical convolution ignores the radial symmetry of the BEV features and increases the difficulty of the detector optimization.

3D Object Detection Depth Estimation +3

EAPruning: Evolutionary Pruning for Vision Transformers and CNNs

no code implementations1 Oct 2022 Qingyuan Li, Bo Zhang, Xiangxiang Chu

In this paper, we undertake a simple and effective approach that can be easily applied to both vision transformers and convolutional neural networks.

Fully Convolutional One-Stage 3D Object Detection on LiDAR Range Images

no code implementations27 May 2022 Zhi Tian, Xiangxiang Chu, Xiaoming Wang, Xiaolin Wei, Chunhua Shen

In this work, we tackle this challenging issue with a novel range view projection mechanism, and for the first time demonstrate the benefits of fusing multi-frame point clouds for a range-view based detector.

3D Object Detection Autonomous Driving +2

PromptDet: Towards Open-vocabulary Detection using Uncurated Images

2 code implementations30 Mar 2022 Chengjian Feng, Yujie Zhong, Zequn Jie, Xiangxiang Chu, Haibing Ren, Xiaolin Wei, Weidi Xie, Lin Ma

The goal of this work is to establish a scalable pipeline for expanding an object detector towards novel/unseen categories, using zero manual annotations.

Language Modelling Object

DAAS: Differentiable Architecture and Augmentation Policy Search

no code implementations30 Sep 2021 Xiaoxing Wang, Xiangxiang Chu, Junchi Yan, Xiaokang Yang

Neural architecture search (NAS) has been an active direction of automatic machine learning (Auto-ML), aiming to explore efficient network structures.

Data Augmentation Neural Architecture Search

Twins: Revisiting the Design of Spatial Attention in Vision Transformers

8 code implementations NeurIPS 2021 Xiangxiang Chu, Zhi Tian, Yuqing Wang, Bo Zhang, Haibing Ren, Xiaolin Wei, Huaxia Xia, Chunhua Shen

Very recently, a variety of vision transformer architectures for dense prediction tasks have been proposed and they show that the design of spatial attention is critical to their success in these tasks.

Image Classification Semantic Segmentation

A Strong On-Policy Competitor To PPO

no code implementations1 Jan 2021 Xiangxiang Chu

As a recognized variant and improvement for Trust Region Policy Optimization (TRPO), proximal policy optimization (PPO) has been widely used with several advantages: efficient data utilization, easy implementation and good parallelism.

A Unified Mixture-View Framework for Unsupervised Representation Learning

no code implementations26 Nov 2020 Xiangxiang Chu, Xiaohang Zhan, Bo Zhang

Recent unsupervised contrastive representation learning follows a Single Instance Multi-view (SIM) paradigm where positive pairs are usually constructed with intra-image data augmentation.

Data Augmentation object-detection +2

ROME: Robustifying Memory-Efficient NAS via Topology Disentanglement and Gradient Accumulation

no code implementations ICCV 2023 Xiaoxing Wang, Xiangxiang Chu, Yuda Fan, Zhexi Zhang, Bo Zhang, Xiaokang Yang, Junchi Yan

Albeit being a prevalent architecture searching approach, differentiable architecture search (DARTS) is largely hindered by its substantial memory cost since the entire supernet resides in the memory.

Disentanglement Neural Architecture Search

Noisy Differentiable Architecture Search

1 code implementation7 May 2020 Xiangxiang Chu, Bo Zhang

However, it largely suffers from the well-known performance collapse issue due to the aggregation of skip connections.

Image Classification Neural Architecture Search

MixPath: A Unified Approach for One-shot Neural Architecture Search

1 code implementation ICCV 2023 Xiangxiang Chu, Shun Lu, Xudong Li, Bo Zhang

However, current two-stage neural architecture search methods are mainly limited to single-path search spaces.

Neural Architecture Search

Neural Architecture Search on Acoustic Scene Classification

no code implementations30 Dec 2019 Jixiang Li, Chuming Liang, Bo Zhang, Zhao Wang, Fei Xiang, Xiangxiang Chu

Convolutional neural networks are widely adopted in Acoustic Scene Classification (ASC) tasks, but they generally carry a heavy computational burden.

Acoustic Scene Classification Classification +3

MoGA: Searching Beyond MobileNetV3

2 code implementations4 Aug 2019 Xiangxiang Chu, Bo Zhang, Ruijun Xu

Bearing the target hardware in mind, we propose the first Mobile GPU-Aware (MoGA) neural architecture search in order to be precisely tailored for real-world applications.

Image Classification Neural Architecture Search

A Matrix-in-matrix Neural Network for Image Super Resolution

1 code implementation19 Mar 2019 Hailong Ma, Xiangxiang Chu, Shaohua Wan, Bo Zhang

In recent years, deep learning methods have achieved impressive results with higher peak signal-to-noise ratio in single image super-resolution (SISR) tasks by utilizing deeper layers.

Image Super-Resolution

Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search

4 code implementations4 Jan 2019 Xiangxiang Chu, Bo Zhang, Ruijun Xu, Hailong Ma

In this paper, we present a new multi-objective oriented algorithm called MoreMNAS (Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search) by leveraging good virtues from both EA and RL.

Image Classification Neural Architecture Search +1

Improved Crowding Distance for NSGA-II

no code implementations30 Nov 2018 Xiangxiang Chu, Xinjie Yu

Non-dominated sorting genetic algorithm II (NSGA-II) does well in dealing with multi-objective problems.

Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization

2 code implementations2 Jul 2018 Xiangxiang Chu

As the most successful variant and improvement for Trust Region Policy Optimization (TRPO), proximal policy optimization (PPO) has been widely applied across various domains with several advantages: efficient data utilization, easy implementation, and good parallelism.

Atari Games

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