检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Kim, Minju Bae, Joonhyeon Shin, Eunsik Lee, Kyogu Department of Intelligence and Information Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Department of Intelligence and Information Interdisciplinary Program in Artificial Intelligence Artificial Intelligence Institute Seoul National University Korea Republic of

ISBN: (纸本)9798350368741

Most studies on music source separation have traditionally concentrated on separating popular music into vocals, bass, drums, and others, with fewer studies focusing on chamber ensemble audio. The task of separating sources from chamber ensemble audio presents greater challenges due to factors like timbral similarity, high synchronization, and spectral overlap. These complexities necessitate more refined and realistic datasets for effective chamber ensemble separation. Yet, the creation of such datasets faces numerous challenges, depending on the methods of dataset formation. Taking these challenges into account, our work seeks to enhance the performance of chamber ensemble separation tasks, with a particular focus on string quartets. Based on dataset generation using a neural synthesis model, we propose an approach that incorporates musical expressions into the dataset generation process using MusicXML files. Furthermore, our framework adapt reverberation to the audio to better match the target mixtures. Our objective evaluation shows an increase in performance of chamber ensemble separation, supported by a subjective listening test to demonstrate improvement in our dataset. We also release our dataset and source code for further usage. © 2025 IEEE.

关键词： Chamber Music Ensemble Separation Music Source Separation

来源：评论

学校读者我要写书评

暂无评论

Regularizing Hard Examples Improves Adversarial Robustness

引用

JOURNAL OF MACHINE LEARNING RESEARCH 2025年 26卷

作者： Lee, Hyungyu Lee, Saehyung Bae, Ho Yoon, Sungroh Seoul Natl Univ Elect & Comp Engn Interdisciplinary Program Artificial Intelligence Seoul 08826 South Korea Ewha Womans Univ Dept Cyber Secur Seoul 03760 South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Elect & Comp Engn AIISASRIINMC Seoul 08826 South Korea Seoul Natl Univ ISRC Seoul 08826 South Korea

Recent studies have validated that pruning hard-to-learn examples from training improves the generalization performance of neural networks (NNs). In this study, we investigate this intriguing phenomenon-the negative effect of hard examples on generalization-in adversarial training. Particularly, we theoretically demonstrate that the increase in the difficulty of hard examples in adversarial training is significantly greater than the increase in the difficulty of easy examples. Furthermore, we verify that hard examples are only fitted through memorization of the label in adversarial training. We conduct both theoretical and empirical analyses of this memorization phenomenon, showing that pruning hard examples in adversarial training can enhance the model's robustness. However, the challenge remains in finding the optimal threshold for removing hard examples that degrade robustness performance. Based upon these observations, we propose a new approach, difficulty proportional label smoothing (DPLS), to adaptively mitigate the negative effect of hard examples, thereby improving the adversarial robustness of NNs. Notably, our experimental result indicates that our method can successfully leverage hard examples while circumventing the negative effect.

关键词： adversarial robustness adversarial examples hard examples memorization robust overfitting

来源：评论

学校读者我要写书评

暂无评论

Role of the upper-stratospheric mean state in the 2021 sudden stratospheric warming onset

引用

CLIMATE DYNAMICS 2025年第4期63卷 1-15页

作者： Cho, Hyeong-Oh Kim, Young-Ha Son, Seok-Woo Seoul Natl Univ Sch Earth & Environm Sci Seoul South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul South Korea

This study investigates the role of upper-stratospheric preconditioning in the 2021 sudden stratospheric warming (SSW) onset by conducting a series of numerical model experiments. A free-run simulation predicts the SSW onset with a lead time of 13 days, in agreement with operational subseasonal-to-seasonal predictions. The limited prediction skill is partly due to misrepresentation of the polar vortex near the stratopause, which results in weaker upward propagation of planetary-scale waves at high latitudes. A sensitivity experiment is performed by nudging the zonal-mean flow in the upper stratosphere toward observations. With the realistic upper-stratospheric mean flow, the model successfully predicts the SSW onset with a lead time of up to 15 days. Although the improvement is marginal, this simulation predicts the SSW onset more robustly than the free-run simulation and realistically reproduces the observed features of planetary-scale wave propagation into the polar region, guided by the poleward shift of the polar vortex in the upper stratosphere. This result provides modeling evidence for the importance of the upper-stratospheric mean state in the 2021 SSW onset and its prediction.

关键词： Sudden stratospheric warming Vortex preconditioning Numerical nudging

来源：评论

学校读者我要写书评

暂无评论

Enhancing SAR ATR Accuracy for Military Targets: A Novel Approach Using XTI and Amplitude Data Integration

引用

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2025年 22卷

作者： An, Sangho Han, Shinhye Kim, Duk-Jin Seoul Natl Univ Sch Earth & Environm Sci Seoul 08826 South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul 08826 South Korea

Many studies on synthetic aperture radar automatic target recognition (SAR ATR) have aimed to enhance recognition accuracy. Using across-track interferometry (XTI) provides height information of targets, and the 3-D structure of targets in 2-D SAR images can be effectively employed to improve target recognition accuracy. Therefore, this study aims to enhance target detection and recognition accuracy by simultaneously using both conventional amplitude images and XTI interferograms from simulation data in training. The research compared target detection and recognition accuracy between training with only single-band amplitude images generated by RaySAR and training with dual-band images combining amplitude images and XTI interferograms. The results showed that dual-band images significantly outperformed single-band images, with an overall 15.53% increase in F_{1}$ score. This advancement enhances the practical applicability of SAR ATR in various conditions and for various targets, especially with the increasing availability of satellites capable of operating XTI.

关键词： Automatic target recognition (ATR) interfer- ometry simulated data synthetic aperture radar (SAR) synthetic aperture radar (SAR) synthetic aperture radar (SAR)

来源：评论

学校读者我要写书评

暂无评论

PERC: Plan-As-Query Example Retrieval for Underrepresented Code Generation 31

PERC: Plan-As-Query Example Retrieval for Underrepresented C...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Yoo, Jaeseok Han, Hojae Lee, Youngwon Kim, Jaejin Hwang, Seung-Won Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of

ISBN: (纸本)9798891761964

Code generation with large language models has shown significant promise, especially when employing retrieval-augmented generation (RAG) with few-shot examples. However, selecting effective examples that enhance generation quality remains a challenging task, particularly when the target programming language (PL) is underrepresented. In this study, we present two key findings: (1) retrieving examples whose presented algorithmic plans can be referenced for generating the desired behavior significantly improves generation accuracy, and (2) converting code into pseudocode effectively captures such algorithmic plans, enhancing retrieval quality even when the source and the target PLs are different. Based on these findings, we propose Plan-as-query Example Retrieval for few-shot prompting in Code generation (PERC), a novel framework that utilizes algorithmic plans to identify and retrieve effective examples. We validate the effectiveness of PERC through extensive experiments on the CodeContests, HumanEval and MultiPL-E benchmarks: PERC consistently outperforms the state-of-the-art RAG methods in code generation, both when the source and target programming languages match or differ, highlighting its adaptability and robustness in diverse coding environments. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Multi-level analyzation of imbalance to resolve non-IID-ness in federated learning

引用

NEUROCOMPUTING 2025年 626卷

作者： Chung, Haengbok Lee, Jae Sung Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul South Korea Seoul Natl Univ Coll Med Dept Nucl Med Seoul South Korea Brightonix Imaging Inc Seoul South Korea

Class imbalance is a common problem in deep learning that severely degrades performance. In federated learning (FL), it is a critical factor contributing to non-identically distributed data (non-IID). Building on several previous attempts, we define and analyze imbalance issues in FL at three levels: inter-case, inter-class, and inter-client. Inter-case imbalance addresses the imbalance in every single class;inter-class imbalance compares the number of data between different classes. Inter-client imbalance represents different skewness of local data between clients. Based on these concepts, we propose FedBB, which consists of two main components: (1) Positive Negative Balanced (PNB) loss function addresses the inter-case and inter-class imbalances in local training, enhancing generalization on highly skewed local client datasets. It optimizes both multi-label and multi-class classifications by assigning higher weights to minority cases or classes. (2) Client Balanced Reweighting (CBR) reweights clients based on inter-client imbalance during model aggregation, giving greater weight to models trained on less skewed datasets. Various experiments on X-ray and natural image datasets demonstrate that FedBB outperforms other algorithms in both performance and efficiency. Additionally, it requires limited statistical information, which is beneficial for privacy protection. Through ablation studies, we proved that PNB loss and CBR independently contribute to performance. As FedBB aims to build a global model that accurately classifies all classes, it can serve as a robust baseline for the generic and personalized FL.

关键词： On-device Federated learning Privacy Security Class imbalance Non-IID

来源：评论

学校读者我要写书评

暂无评论

Moment-Aware Video Retrieval for Video Corpus Moment Retrieval

引用

IEEE ACCESS 2025年 13卷 38574-38583页

作者： Choi, Yura Ahn, Daechul Choi, Jonghyun Seoul Natl Univ Dept Elect & Comp Engn Seoul 08733 South Korea Yonsei Univ Dept Artificial Intelligence Seoul 03788 South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul 08733 South Korea Seoul Natl Univ Automat & Syst Res Inst Seoul 08733 South Korea

Video corpus moment retrieval (VCMR) task aims to retrieve a specific moment from a large corpus of untrimmed videos. This task has been addressed by decomposing it into video retrieval and moment retrieval subtasks each with specialized heads (i.e., democratic decomposition), due to the computational complexity. However, this approach overlooks the interdependency between the subtasks, which is crucial for temporally fine-grained query-video alignment. To address this suboptimality, we propose an integrated learning framework that explicitly establishes a connection between the two subtasks, namely moment-aware video retrieval, allowing them to benefit from each other's learning process. Furthermore, we employ a curriculum-based negative sampling strategy that gradually provides harder negative samples to enhance the discriminative ability between semantically similar negative videos. We empirically show that the proposed approach outperforms state-of-the-art methods on three benchmarks - TVR, ActivityNet, and DiDeMo. Notably, on TVR, our method achieves 10.04% in VCMR R@1 with tIoU=0.7, representing a 1.68% absolute improvement over prior work, and similarly demonstrates consistent gains on ActivityNet (4.98% in VCMR R@1 with tIoU=0.5) and DiDeMo, validating the effectiveness of our integrated approach.

关键词： Training Semantics Visualization Predictive models Medical services Benchmark testing Text to video Object recognition Bars Solid modeling Multi-modal learning temporal moment retrieval video corpus moment retrieval video retrieval video-text alignment

来源：评论

学校读者我要写书评

暂无评论

TokenSynth: A Token-based Neural Synthesizer for Instrument Cloning and Text-to-Instrument

TokenSynth: A Token-based Neural Synthesizer for Instrument ...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Kim, Kyungsu Koo, Junghyun Lee, Sungho Joung, Haesun Lee, Kyogu Department of Intelligence and Information Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Artificial Intelligence Institute Seoul National University Korea Republic of

ISBN: (纸本)9798350368741

Recent advancements in neural audio codecs have enabled the use of tokenized audio representations in various audio generation tasks, such as text-to-speech, text-to-audio, and text-to-music generation. Leveraging this approach, we propose TokenSynth, a novel neural synthesizer that utilizes a decoder-only transformer to generate desired audio tokens from MIDI tokens and CLAP (Contrastive Language-Audio Pretraining) embedding, which has timbre-related information. Our model is capable of performing instrument cloning, text-to-instrument synthesis, and text-guided timbre manipulation without any fine-tuning. This flexibility enables diverse sound design and intuitive timbre control. We evaluated the quality of the synthesized audio, the timbral similarity between synthesized and target audio/text, and synthesis accuracy (i.e., how accurately it follows the input MIDI) using objective *** demonstrates the potential of leveraging advanced neural audio codecs and transformers to create powerful and versatile neural synthesizers. The source code, model weights, and audio demos are available at: https://***/KyungsuKim42/tokensynth. © 2025 IEEE.

关键词： instrument cloning MIDI neural audio codec Neural synthesizer text-to-instrument transformer

来源：评论

学校读者我要写书评

暂无评论

On Mitigating Stability-Plasticity Dilemma in CLIP-guided Image Morphing via Geodesic Distillation Loss

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2025年第5期133卷 2721-2751页

作者： Oh, Yeongtak Lee, Saehyung Hwang, Uiwon Yoon, Sungroh Seoul Natl Univ Dept Elect & Comp Engn Seoul 08826 South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul 08826 South Korea Yonsei Univ Div Digital Healthcare Wonju 26493 South Korea

Large-scale language-vision pre-training models, such as CLIP, have achieved remarkable results in text-guided image morphing by leveraging several unconditional generative models. However, existing CLIP-guided methods face challenges in achieving photorealistic morphing when adapting the generator from the source to the target domain. Specifically, current guidance methods fail to provide detailed explanations of the morphing regions within the image, leading to misguidance and catastrophic forgetting of the original image's fidelity. In this paper, we propose a novel approach considering proper regularization losses to overcome these difficulties by addressing the SP dilemma in CLIP guidance. Our approach consists of two key components: (1) a geodesic cosine similarity loss that minimizes inter-modality features (i.e., image and text) in a projected subspace of CLIP space, and (2) a latent regularization loss that minimizes intra-modality features (i.e., image and image) on the image manifold. By replacing the naive directional CLIP loss in a drop-in replacement manner, our method achieves superior morphing results for both images and videos across various benchmarks, including CLIP-inversion.

关键词： Image morphing CLIP Stability-plasticity dilemma Geodesic path

来源：评论

学校读者我要写书评

暂无评论

The Effects of Musical Factors on the Perception of Auditory Illusions

引用

TOPICS IN COGNITIVE SCIENCE 2025年第1期17卷 106-119页

作者： Choi, Ahyeon Bang, Younyoung Park, Jeong Mi Lee, Kyogu Seoul Natl Univ Dept Intelligence & Informat 1 Gwanak Ro Seoul 08826 South Korea Seoul Natl Univ Dept Musicol Seoul South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul South Korea Seoul Natl Univ Artificial Intelligence Inst Seoul South Korea

This study delves into how various musical factors influence the experience of auditory illusions, building on Diana Deutsch's scale illusion experiments and subsequent studies. Exploring the interaction between scale mode and timbre, this study assesses their influence on auditory misperceptions, while also considering the impact of an individual's musical training and ability to discern absolute pitch. Participants were divided into nonmusicians, musicians with absolute pitch, and musicians with relative pitch, and were exposed to stimuli modified across three scale modes (tonal, dissonant, atonal) and two timbres (same, different). The findings suggest that scale illusions occur less frequently with different timbres and vary with scale mode. Crucially, the absolute pitch ability appears to have a more significant impact on the perception of illusions than the duration of musical training. This research contributes to understanding the complex interplay between various factors in auditory perception and the mechanisms behind the experience of auditory illusions.

关键词： Auditory illusion Scale illusion Timbre Mode Musical training Absolute pitch

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：