检索结果-内蒙古大学图书馆

Bad Actor, Good Advisor: Exploring the Role of Large Language Models in Fake News Detection

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hu, Beizhe Sheng, Qiang Cao, Juan Shi, Yuhui Li, Yang Wang, Danding Qi, Peng Key Lab of Intelligent Information Processing Chinese Academy of Sciences Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China National University of Singapore Singapore

Detecting fake news requires both a delicate sense of diverse clues and a profound understanding of the real-world background, which remains challenging for detectors based on small language models (SLMs) due to their knowledge and capability limitations. Recent advances in large language models (LLMs) have shown remarkable performance in various tasks, but whether and how LLMs could help with fake news detection remains underexplored. In this paper, we investigate the potential of LLMs in fake news detection. First, we conduct an empirical study and find that a sophisticated LLM such as GPT 3.5 could generally expose fake news and provide desirable multi-perspective rationales but still underperforms the basic SLM, fine-tuned BERT. Our subsequent analysis attributes such a gap to the LLM’s inability to select and integrate rationales properly to conclude. Based on these findings, we propose that current LLMs may not substitute fine-tuned SLMs in fake news detection but can be a good advisor for SLMs by providing multi-perspective instructive rationales. To instantiate this proposal, we design an adaptive rationale guidance network for fake news detection (ARG), in which SLMs selectively acquire insights on news analysis from the LLMs’ rationales. We further derive a rationale-free version of ARG by distillation, namely ARG-D, which services cost-sensitive scenarios without querying LLMs. Experiments on two real-world datasets demonstrate that ARG and ARG-D outperform three types of baseline methods, including SLM-based, LLM-based, and combinations of small and large language models. © 2023, CC BY-NC-SA.

关键词： Distillation

AdaDiff: Adaptive Step Selection for Fast Diffusion

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Hui Wu, Zuxuan Xing, Zhen Shao, Jie Jiang, Yu-Gang Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China Shanghai Collaborative Innovation Center of Intelligent Visual Computing China ByteDance Inc China

Diffusion models, as a type of generative models, have achieved impressive results in generating images and videos conditioned on textual conditions. However, the generation process of diffusion models involves denoising for dozens of steps to produce photorealistic images/videos, which is computationally expensive. Unlike previous methods that design "one-size-fits-all" approaches for speed up, we argue denoising steps should be sample-specific conditioned on the richness of input texts. To this end, we introduce AdaDiff, a lightweight framework designed to learn instance-specific step usage policies, which are then used by the diffusion model for generation. AdaDiff is optimized using a policy gradient method to maximize a carefully designed reward function, balancing inference time and generation quality. We conduct experiments on three image generation and two video generation benchmarks and demonstrate that our approach achieves similar results in terms of visual quality compared to the baseline using a fixed 50 denoising steps while reducing inference time by at least 33%, going as high as 40%. Furthermore, our qualitative analysis shows that our method allocates more steps to more informative text conditions and fewer steps to simpler text conditions. Copyright © 2023, The Authors. All rights reserved.

关键词： Benchmarking

Edge-Preserving Multi-Exposure Fusion with Guided Filtering for Structural Patch Decomposition

学校读者我要写书评

暂无评论

Edge-Preserving Multi-Exposure Fusion with Guided Filtering ...

IEEE International Conference on Electronic Information and Communication Technology (ICEICT)

作者： Yang Zhen Junwei Qi Yingsong Li College of Information and Communication Engineering Harbin Engineering University Harbin China College of Computer Science and Technology Harbin Engineering University Harbin China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei China

Multi-exposure image fusion (MEF) is one of the effective ways to obtain high dynamic range (HDR) images. Multi-scale structural patch decomposition based MEF(MSPD) and edge-preserving structural patch decomposition (MESPD) have achieved state-of-the-art fusion quality. MESPD incorporates edge-preserving factors into MSPD to preserve details in a single-scale setting. However, we find that the edge-preserving factors put too much emphasis on the edge, resulting in artifacts on some edges and loss of details. To tackle this issue, guided filter is introduced into MSPD, which is directly used as an edge-preserving filter instead of mean filter to implicitly decompose images. In addition, we have designed a flexible adaptive weight, which can further maintain details of the light and dark areas. Likewise, we inherit MESPD and MSPD to extend our method to multi-scale framework. Extensive experimental results demonstrate our method can produce satisfactory results in both static and dynamic scenes with fewer artifacts and low computational cost.

关键词：

Spatio-Temporal Coherent Snapshot Compressed Imaging with Time-Varying Composite Phase Masks

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Zhang, Ru Zhang, Cheng Zhang, Liru Chen, Mingsheng Cheng, Hong Shen, Chuan Zhang, Quanbin Wei, Sui Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Anhui Province Hefei230601 China Advanced Laser Technology Laboratory of Anhui Province National University of Defense Technology Anhui Province Hefei230037 China

Snapshot compressed imaging (SCI) can encode an image sequence into a single-frame measurement and has the advantages of low requirements for physical sensors, low bandwidth/ memory requirements, and a fast acquisition speed. However, existing snapshot compressed imaging mainly focuses on incoherent imaging, such as video/hyperspectral imaging, and research on coherent imaging is rare. In this paper, snapshot compressed sensing is introduced for coherent imaging conditions, and a spatio-temporal coherent snapshot compressed imaging method with time-varying composite phase masks is proposed. Numerical experiments demonstrate the effectiveness and superiority of our proposed spatio-temporal coherent snapshot compressed imaging with time-varying composite phase masks method (CSCI-CPM) in terms of the reconstruction quality and robustness for both self-sparse and complex cases. © 2023, The Authors. All rights reserved.

关键词： Numerical methods

Instance-aware Multi-Camera 3D Object Detection with Structural Priors Mining and Self-Boosting Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jiao, Yang Jie, Zequn Chen, Shaoxiang Cheng, Lechao Chen, Jingjing Ma, Lin Jiang, Yu-Gang Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China Shanghai Collaborative Innovation Center on Intelligent Visual Computing China Meituan China Zhejiang Lab. China

Camera-based bird-eye-view (BEV) perception paradigm has made significant progress in the autonomous driving field. Under such a paradigm, accurate BEV representation construction relies on reliable depth estimation for multi-camera images. However, existing approaches exhaustively predict depths for every pixel without prioritizing objects, which are precisely the entities requiring detection in the 3D space. To this end, we propose IA-BEV, which integrates image-plane instance awareness into the depth estimation process within a BEV-based detector. First, a category-specific structural priors mining approach is proposed for enhancing the efficacy of monocular depth generation. Besides, a self-boosting learning strategy is further proposed to encourage the model to place more emphasis on challenging objects in computation-expensive temporal stereo matching. Together they provide advanced depth estimation results for high-quality BEV features construction, benefiting the ultimate 3D detection. The proposed method achieves state-of-the-art performances on the challenging nuScenes benchmark, and extensive experimental results demonstrate the effectiveness of our designs. © 2023, CC BY-NC-SA.

关键词： Object detection

Critical Node Detection for Sequential Attacks in Complex Networks via Multi-objective Optimization 2

学校读者我要写书评

暂无评论

Critical Node Detection for Sequential Attacks in Complex Ne...

2nd International Conference on Industrial Artificial Intelligence, IAI 2020

作者： Zhang, Lei Liu, Qing Xia, Jiajun Yang, Haipeng Zhang, Xingyi Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University China

ISBN: (纸本)9781728182162

The critical node detection for sequential attacks based on cascading failure model is an important way for analyzing network vulnerability, which has attracted the attention of many researchers in the field of complex network recently. However, most of the existing cascading critical node detection algorithms focus on designing effective attack strategies leading to the maximal damage to the network (i.e. attack effect), while ignoring the cost of attacks. To this end, we transform the cascading critical node detection for sequential attacks as a bi-objective optimization problem (named BCVNDSeq), where the attack cost and the attack effect are simultaneously optimized. In order to solve the transformed problem, we propose a multiobjective cascading critical node detection algorithm (named MO-BCVNDSeq), which can provide decision makers with a holistic view for analyzing the network vulnerability. In MO-BCVNDSeq, a local search strategy based on sequential matrix is proposed to accelerate the population convergence and an individual repairing strategy is also suggested to further improve the search efficiency. Finally, the experimental results on 6 real-world complex networks demonstrate the effectiveness of the proposed algorithm compared with several representative baselines. © 2020 IEEE.

关键词： Multiobjective optimization

AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Han, Boyu Xu, Qianqian Yang, Zhiyong Bao, Shilong Wen, Peisong Jiang, Yangbangyan Huang, Qingming Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China School of Computer Science and Tech. University of Chinese Academy of Sciences China Peng Cheng Laboratory China Key Laboratory of Big Data Mining and Knowledge Management CAS China

The Area Under the ROC Curve (AUC) is a well-known metric for evaluating instance-level long-tail learning problems. In the past two decades, many AUC optimization methods have been proposed to improve model performance under long-tail distributions. In this paper, we explore AUC optimization methods in the context of pixel-level long-tail semantic segmentation, a much more complicated scenario. This task introduces two major challenges for AUC optimization techniques. On one hand, AUC optimization in a pixel-level task involves complex coupling across loss terms, with structured inner-image and pairwise inter-image dependencies, complicating theoretical analysis. On the other hand, we find that mini-batch estimation of AUC loss in this case requires a larger batch size, resulting in an unaffordable space complexity. To address these issues, we develop a pixel-level AUC loss function and conduct a dependency-graph-based theoretical analysis of the algorithm’s generalization ability. Additionally, we design a Tail-Classes Memory Bank (T-Memory Bank) to manage the significant memory demand. Finally, comprehensive experiments across various benchmarks confirm the effectiveness of our proposed AUCSeg method. The code is available at https://***/boyuh/AUCSeg. © 2024, CC BY.

关键词： Semantic Segmentation

Data Fusion Based on Temporal-Spatial Composite Sequential Three-Way Decision

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Xu, Yi Qiu, Ziheng Sun, Xiaojun Department of Computer Science and Technology Anhui University Hefei230601 China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei230601 China

Three-way decision is the representative of trisecting-acting-outcome(TAO) model. Considering the temporality, in dynamic information systems, the objects may be inserted or deleted,attributes may be added or removed, and attributes values may be updated or modified. These variations can be reflected by the TAO model, which cares about consideration at different levels, acquisition of strategies, and handling of issues based on Strategies. Considering the spatiality, different levels, and different perspectives of granular structure are created by the sequential approach of three-way decision. In this paper, in order to fuse hybrid data, a sequential three-way decision is proposed by the combination of temporality and spatiality. Firstly, We present the frameworks of spatiality-based and temporality-based and focus on temporal-spatial three-way decision. Then we provide three operations to construct the composite POS structure by combining a family of approximations. Based on the composite POS structure, three different strategies for the fusion of dynamic hybrid data are adopted. Finally, we design the algorithm and implement a series of comparison experiments to verify our model is effective. © 2022, The Authors. All rights reserved.

关键词： Data fusion

Look Before You Match: Instance Understanding Matters in Video Object Segmentation

学校读者我要写书评

暂无评论

Look Before You Match: Instance Understanding Matters in Vid...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Junke Wang Dongdong Chen Zuxuan Wu Chong Luo Chuanxin Tang Xiyang Dai Yucheng Zhao Yujia Xie Lu Yuan Yu-Gang Jiang Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University Shanghai Collaborative Innovation Center of Intelligent Visual Computing Microsoft Cloud + AI Microsoft Research Asia

Exploring dense matching between the current frame and past frames for long-range context modeling, memory-based methods have demonstrated impressive results in video object segmentation (VOS) recently. Nevertheless, due to the lack of instance understanding ability, the above approaches are oftentimes brittle to large appearance variations or viewpoint changes resulted from the movement of objects and cameras. In this paper, we argue that instance understanding matters in VOS, and integrating it with memory-based matching can enjoy the synergy, which is intuitively sensible from the definition of VOS task, i.e., identifying and segmenting object instances within the video. Towards this goal, we present a two-branch network for VOS, where the query-based instance segmentation (IS) branch delves into the instance details of the current frame and the VOS branch performs spatial-temporal matching with the memory bank. We employ the well-learned object queries from IS branch to inject instance-specific information into the query key, with which the instance-augmented matching is further performed. In addition, we introduce a multi-path fusion block to effectively combine the memory readout with multi-scale features from the instance segmentation decoder, which incorporates high-resolution instance-aware features to produce final segmentation results. Our method achieves state-of-the-art performance on DAVIS 2016/2017 val (92.6% and 87.1%), DAVIS 2017 test-dev (82.8%), and YouTube-VOS 2018/2019 val (86.3% and 86.3%), outperforming alternative methods by clear margins.

关键词：