检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Shen, Siqi Ma, Chennan Li, Chao Liu, Weiquan Fu, Yongquan Mei, Songzhu Liu, Xinwang Wang, Cheng China Key Laboratory of Multimedia Trusted Perception and Efficient Computing XMU China School of Computer National University of Defense Technology China

Multi-agent systems are characterized by environmental uncertainty, varying policies of agents, and partial observability, which result in significant risks. In the context of Multi-Agent Reinforcement Learning (MARL), learning coordinated and decentralized policies that are sensitive to risk is challenging. To formulate the coordination requirements in risk-sensitive MARL, we introduce the Risk-sensitive Individual-Global-Max (RIGM) principle as a generalization of the Individual-Global-Max (IGM) and Distributional IGM (DIGM) principles. This principle requires that the collection of risk-sensitive action selections of each agent should be equivalent to the risk-sensitive action selection of the central policy. Current MARL value factorization methods do not satisfy the RIGM principle for common risk metrics such as the Value at Risk (VaR) metric or distorted risk measurements. Therefore, we propose RiskQ to address this limitation, which models the joint return distribution by modeling quantiles of it as weighted quantile mixtures of per-agent return distribution utilities. RiskQ satisfies the RIGM principle for the VaR and distorted risk metrics. We show that RiskQ can obtain promising performance through extensive experiments. The source code of RiskQ is available in https://***/xmu-rl-3dv/RiskQ. Copyright © 2023, The Authors. All rights reserved.

关键词： Value engineering

来源：评论

学校读者我要写书评

暂无评论

FlashSloth: Lightning Multimodal Large Language Models via Embedded Visual Compression

arXiv

引用

arXiv 2024年

作者： Tong, Bo Lai, Bokai Zhou, Yiyi Luo, Gen Shen, Yunhang Li, Ke Sun, Xiaoshuai Ji, Rongrong Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University 361005 China Youtu Lab Tencent China OpenGVLab Shanghai AI Laboratory China

Despite a big leap forward in capability, multimodal large language models (MLLMs) tend to behave like a sloth in practical use, i.e., slow response and large latency. Recent efforts are devoted to building tiny MLLMs for better efficiency, but the plethora of visual tokens still used limit their actual speedup. In this paper, we propose a powerful and fast tiny MLLM called FlashSloth. Different from previous efforts, FlashSloth focuses on improving the descriptive power of visual tokens in the process of compressing their redundant semantics. In particular, FlashSloth introduces embedded visual compression designs to capture both visually salient and instruction-related image information, so as to achieving superior multimodal performance with fewer visual tokens. Extensive experiments are conducted to validate the proposed FlashSloth, and a bunch of tiny but strong MLLMs are also comprehensively compared, e.g., InternVL2, MiniCPM-V2 and Qwen2-VL. The experimental results show that compared with these advanced tiny MLLMs, our FlashSloth can greatly reduce the number of visual tokens, training memory and computation complexity while retaining high performance on various VL tasks. Our code is released at: https://***/codefanw/FlashSloth. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Diffree: Text-Guided Shape Free Object Inpainting with Diffusion Model

arXiv

引用

arXiv 2024年

作者： Zhao, Lirui Yang, Tianshuo Shao, Wenqi Zhang, Yuxin Qiao, Yu Luo, Ping Zhang, Kaipeng Ji, Rongrong Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University China OpenGVLab Shanghai AI Laboratory China The University of Hong Kong Hong Kong

This paper addresses an important problem of object addition for images with only text guidance. It is challenging because the new object must be integrated seamlessly into the image with consistent visual context, such as lighting, texture, and spatial location. While existing text-guided image inpainting methods can add objects, they either fail to preserve the background consistency or involve cumbersome human intervention in specifying bounding boxes or user-scribbled masks. To tackle this challenge, we introduce Diffree, a Text-to-Image (T2I) model that facilitates text-guided object addition with only text control. To this end, we curate OABench, an exquisite synthetic dataset by removing objects with advanced image inpainting techniques. OABench comprises 74K real-world tuples of an original image, an inpainted image with the object removed, an object mask, and object descriptions. Trained on OABench using the Stable Diffusion model with an additional mask prediction module, Diffree uniquely predicts the position of the new object and achieves object addition with guidance from only text. Extensive experiments demonstrate that Diffree excels in adding new objects with a high success rate while maintaining background consistency, spatial appropriateness, and object relevance and quality. © 2024, CC BY.

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

MMICT: Boosting Multi-Modal Fine-Tuning with In-Context Examples

arXiv

引用

arXiv 2023年

作者： Chen, Tao Zhang, Enwei Gao, Yuting Li, Ke Sun, Xing Zhang, Yan Li, Hui Ji, Rongrong Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University China Tencent Youtu Lab China

Although In-Context Learning (ICL) brings remarkable performance gains to Large Language Models (LLMs), the improvements remain lower than fine-tuning on downstream tasks. This paper introduces Multi-Modal In-Context Tuning (MMICT), a novel multi-modal fine-tuning paradigm that boosts multi-modal fine-tuning by fully leveraging the promising ICL capability of multi-modal LLMs (MM-LLMs). We propose the Multi-Modal Hub (M-Hub), a unified module that captures various multi-modal features according to different inputs and objectives. Based on M-Hub, MMICT enables MM-LLMs to learn from in-context visual-guided textual features and subsequently generate outputs conditioned on the textual-guided visual features. Moreover, leveraging the flexibility of M-Hub, we design a variety of in-context demonstrations. Extensive experiments on a diverse range of downstream multi-modal tasks demonstrate that MMICT significantly outperforms traditional fine-tuning strategy and the vanilla ICT method that directly takes the concatenation of all information from different modalities as input. Our implementation is available at: https://***/KDEGroup/MMICT. Copyright © 2023, The Authors. All rights reserved.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

SMMix: Self-Motivated Image Mixing for Vision Transformers

SMMix: Self-Motivated Image Mixing for Vision Transformers

引用

International Conference on Computer Vision (ICCV)

作者： Mengzhao Chen Mingbao Lin Zhihang Lin Yuxin Zhang Fei Chao Rongrong Ji Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China School of Informatics Xiamen University Tencent Youtu Lab

CutMix is a vital augmentation strategy that determines the performance and generalization ability of vision transformers (ViTs). However, the inconsistency between the mixed images and the corresponding labels harms its efficacy. Existing CutMix variants tackle this problem by generating more consistent mixed images or more precise mixed labels, but inevitably introduce heavy training overhead or require extra information, undermining ease of use. To this end, we propose an novel and effective Self-Motivated image Mixing method (SMMix), which motivates both image and label enhancement by the model under training itself. Specifically, we propose a max-min attention region mixing approach that enriches the attention-focused objects in the mixed images. Then, we introduce a fine-grained label assignment technique that co-trains the output tokens of mixed images with fine-grained supervision. Moreover, we devise a novel feature consistency constraint to align features from mixed and unmixed images. Due to the subtle designs of the self-motivated paradigm, our SMMix is significant in its smaller training overhead and better performance than other CutMix variants. In particular, SMMix improves the accuracy of DeiT-T/S/B, CaiT-XXS-24/36, and PVT-T/S/M/L by more than +1% on ImageNet-1k. The generalization capability of our method is also demonstrated on downstream tasks and out-of-distribution datasets. Our project is available at https://***/ChenMnZ/SMMix.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FocSAM: Delving Deeply into Focused Objects in Segmenting Anything

FocSAM: Delving Deeply into Focused Objects in Segmenting An...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： You Huang Zongyu Lan Liujuan Cao Xianming Lin Shengchuan Zhang Guannan Jiang Rongrong Ji Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University Intelligent Manufacturing Department Contemporary Amperex Technology Co. Limited (CATL)

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

The Segment Anything Model (SAM) marks a notable milestone in segmentation models, highlighted by its robust zero-shot capabilities and ability to handle diverse prompts. SAM follows a pipeline that separates interactive segmentation into image preprocessing through a large encoder and interactive inference via a lightweight decoder, ensuring efficient real-time performance. However, SAM faces stability issues in challenging samples upon this pipeline. These issues arise from two main factors. Firstly, the image preprocessing disables SAM to dynamically use image-level zoom-in strategies to refocus on the target object during interaction. Secondly, the lightweight decoder struggles to sufficiently integrate interactive information with image embeddings. To address these two limitations, we propose FocSAM with a pipeline redesigned on two pivotal aspects. First, we propose Dynamic Window Multi-head Self-Attention (Dwin-MSA) to dynamically refocus SAM's image embeddings on the target object. Dwin-MSA localizes attention computations around the target object, enhancing object-related embeddings with minimal computational overhead. Second, we propose Pixel-wise Dynamic ReLU (P-DyReLU) to enable sufficient integration of interactive information from a few initial clicks that have significant impacts on the overall segmentation results. Experimentally, FocSAM augments SAM's interactive segmentation performance to match the existing state-of-the-art method in segmentation quality, requiring only about 5.6% of this method's inference time on CPUs. Code is available at https://***/YouHuang67/focsam.

关键词： Image segmentation Computer vision Codes Pipelines Stability analysis Real-time systems Decoding

来源：评论

学校读者我要写书评

暂无评论

DuPI: Dual-resolution Pseudo-label Integration for Semi-supervised Instance Segmentation

DuPI: Dual-resolution Pseudo-label Integration for Semi-supe...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yue Ma Jie Hu Chen Chen Shengchuan Zhang Xianming Lin Liujuan Cao Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University P.R. China Learning and Vision Lab National University of Singapore Singapore

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

The role of high-quality pseudo-labels is pivotal in semi-supervised instance segmentation (SSIS). However, existing SSIS frameworks predominantly produce pseudo-labels at a single resolution, which can introduce noise that adversely affects the quality of learning at both the pixel level and in terms of class discrimination. This paper introduces the Dual-Resolution Pseudo-Label Integration for Semi-Supervised Instance Segmentation (DuPI), a novel framework designed to enhance learning by integrating pseudo-labels derived from dual-resolution inputs. The DuPI framework incorporates a Dual-Resolution Pseudo-Label Correction (DPC) module, which refines pseudo-labels through a process of cross-resolution rectification and fusion. Furthermore, the framework introduces an Area-Adaptive Learning (AAL) strategy aimed at enhancing the quality of pseudo-labels sourced from extra-resolution inputs. The AAL strategy addresses the training challenges associated with small objects at lower resolutions by re-weighting pseudo-labels corresponding to tiny mask areas using Intersection over Union (IoU) metrics from the assignments. Experiments on the COCO and BDD100K datasets demonstrate that DuPI achieves state-of-the-art SSIS performance under various semi-supervised settings.

关键词： Instance segmentation Training Measurement Noise Pipelines Acoustics Speech processing Signal resolution

来源：评论

学校读者我要写书评

暂无评论

DYNAMIC SPARSE NO TRAINING ⊘: TRAINING-FREE FINE-TUNING FOR SPARSE LLMS 12

DYNAMIC SPARSE NO TRAINING ⊘: TRAINING-FREE FINE-TUNING FOR...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Zhang, Yuxin Zhao, Lirui Lin, Mingbao Sun, Yunyun Yao, Yiwu Han, Xingjia Tanner, Jared Liu, Shiwei Ji, Rongrong Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University China Pengcheng Lab China Tencent Youtu Lab China Huawei Technologies China University of Oxford United Kingdom University of Texas Austin United States Eindhoven University of Technology Netherlands Institute of Artificial Intelligence Xiamen University China

The ever-increasing large language models (LLMs), though opening a potential path for the upcoming artificial general intelligence, sadly drops a daunting obstacle on the way towards their on-device deployment. As one of the most well-established pre-LLMs approaches in reducing model complexity, network pruning appears to lag behind in the era of LLMs, due mostly to its costly fine-tuning (or re-training) necessity under the massive volumes of model parameter and training data. To close this industry-academia gap, we introduce Dynamic Sparse No Training (DS⊘T), a training-free fine-tuning approach that slightly updates sparse LLMs without the expensive backpropagation and any weight updates. Inspired by the Dynamic Sparse Training, DS⊘T minimizes the reconstruction error between the dense and sparse LLMs, in the fashion of performing iterative weight pruning-and-growing on top of sparse LLMs. To accomplish this purpose, DS⊘T particularly takes into account the anticipated reduction in reconstruction error for pruning and growing, as well as the variance w.r.t. different input data for growing each weight. This practice can be executed efficiently in linear time since its obviates the need of backpropagation for fine-tuning LLMs. Extensive experiments on LLaMA-V1/V2, Vicuna, and OPT across various benchmarks demonstrate the effectiveness of DS⊘T in enhancing the performance of sparse LLMs, especially at high sparsity levels. For instance, DS⊘T is able to outperform the state-of-the-art Wanda by 26.79 perplexity at 70% sparsity with LLaMA-7B. Our paper offers fresh insights into how to fine-tune sparse LLMs in an efficient training-free manner and open new venues to scale the great potential of sparsity to LLMs. Codes are available at https://***/zyxxmu/DSnoT. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Backpropagation

来源：评论

学校读者我要写书评

暂无评论

STGC-NeRF: Spatial-Temporal Geometric Consistency for LiDAR Neural Radiance Fields in Dynamic Scenes 39

STGC-NeRF: Spatial-Temporal Geometric Consistency for LiDAR ...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Yu, Shangshu Sun, Xiaotian Li, Wen Xu, Qingshan Yuan, Zhimin Wang, Sijie She, Rui Wang, Cheng Nanyang Technological University Singapore Fujian Key Laboratory of Sensing and Computing for Smart Cities Xiamen University China Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University China Beihang University China

ISBN: (纸本)157735897X

While Neural Radiance Fields (NeRFs) have advanced the frontiers of novel view synthesis (NVS) using LiDAR data, they still struggle in dynamic scenes. Due to the low frequency and sparsity characteristics of LiDAR point clouds, it is challenging to spontaneously learn a dynamic and consistent scene representation from posed scans. In this paper, we propose STGC-NeRF, a novel LiDAR NeRF method that combines spatial-temporal geometry consistency to enhance the reconstruction of dynamic scenes. First, we propose a temporal geometry consistency regularization to enhance the regression of time-varying scene geometries from low-frequency LiDAR sequences. By estimating the pointwise correspondences between synthetic (or real) and real frames at different times, we convert them into various forms of temporal supervision. This alleviates the inconsistency caused by moving objects in dynamic scenes. Second, to improve the reconstruction of sparse LiDAR data, we propose spatial geometric consistency constraints. By computing multiple neighborhood feature descriptors incorporating geometric and contextual information, we capture structural geometry information from sparse LiDAR data. This helps encourage consistent direction, smoothness, and detail of the local surface. Extensive experiments on the KITTI-360 and nuScenes datasets demonstrate that STGC-NeRF outperforms state-of-the-art methods in both geometry and intensity accuracy for dynamic LiDAR scene reconstruction. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning Based Jamming Detection for Reliable Wireless Communications

Reinforcement Learning Based Jamming Detection for Reliable ...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Chen Wang Yifan Chen Zhiping Lin Qiaoxin Chen Liang Xiao Department of Information and Communication Engineering Xiamen University Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University P.R. China

ISBN: (数字)9798350387414

ISBN: (纸本)9798350387421

Both the radio spectrum features such as power spectral density (PSD) and the communication performance such as packet loss rate (PLR) can be exploited to detect jamming attacks, with the resulting detection results used to enhance the reliability of wireless communications. In this paper, we propose a reinforcement learning (RL)-based jamming detection scheme based on the channel energy, the received signal strength indi-cator of each packet, the channel gains, PLR and transmission latency of mobile devices, in which the test threshold and the number of PSD bins are optimized by access point to enhance the utility as a weighted function of the detection speed and accuracy. The detection results are exploited for mobile devices to choose the transmit power and channel to reduce the PLR and transmission latency. Experimental results based on the universal software radio peripheral and Raspberry Pi to detect four jamming types including constant, sweeping, random and smart jamming show that our proposed schemes improve the detection accuracy and speed, as well as the communication performance.

关键词： Wireless communication Vehicular and wireless technologies Accuracy Channel estimation Packet loss Reinforcement learning Mobile handsets

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：