arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

期刊&会议

NeurIPS

Conference on Neural Information Processing Systems · 会议 · Machine Learning

2026-04-01 至 2026-04-01 共收录 5
2603.28824 2026-04-01 cs.CR cs.AI

SNEAKDOOR: Stealthy Backdoor Attacks against Distribution Matching-based Dataset Condensation

SNEAKDOOR:针对基于分布匹配的数据集压缩的隐秘后门攻击

He Yang, Dongyi Lv, Song Ma, Wei Xi, Jizhong Zhao

机构 * School of Computer Science and Technology, Xi’an Jiaotong University(西安交通大学计算机科学与技术学院) National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi’an Jiaotong University(西安交通大学人机混合增强智能全国重点实验室)

AI总结 本文提出SNEAKDOOR,通过利用类别决策边界漏洞和生成模块,提升隐秘性而不影响攻击效果,实验表明其在攻击成功率、清洁测试准确率和隐秘性之间取得良好平衡。

Comments 29 pages, 5 figures, accepted to NeurIPS 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.20155 2026-04-01 cs.CV

PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding

PartNeXt:面向细粒度和层次化3D部件理解的下一代数据集

Penghao Wang, Yiyang He, Xin Lv, Yukai Zhou, Lan Xu, Jingyi Yu, Jiayuan Gu

机构 * ShanghaiTech University(上海科技大学)

AI总结 PartNeXt通过23000个高质量纹理3D模型,解决传统数据集在可扩展性和实用性上的不足,提升细粒度部件分割和3D部件问答任务的性能。

Comments NeurIPS 2025 DB Track. Project page: https://authoritywang.github.io/partnext

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.12405 2026-04-01 cs.LG cond-mat.mtrl-sci

Continuous SUN (Stable, Unique, and Novel) Metric for Generative Modeling of Inorganic Crystals

连续SUN(稳定、唯一、新颖)度量用于无机晶体生成建模

Masahiro Negishi, Hyunsoo Park, Kinga O. Mastej, Aron Walsh

机构 * Imperial College London(伦敦帝国学院)

AI总结 本文提出连续SUN度量,解决传统二元指标在晶体生成中的局限性,通过连续化提升评估精度与可调性,实验表明其能有效识别优质样本并避免强化学习中的奖励黑客问题。

Comments 23 pages (17 pages of main text). See https://github.com/WMD-group/xtalmet for the code. Significantly extended from the early version of this work, which was accepted to the AI4Mat workshop at NeurIPS 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.19431 2026-04-01 hep-ph cs.LG hep-ex

The Pareto Frontier of Resilient Jet Tagging

可抵御的喷注标记的帕累托前沿

Rikab Gambhir, Matt LeBlanc, Yuanchen Zhou

机构 * University of Cincinnati(辛辛那提大学) IAIFI Brown University(布朗大学)

AI总结 本文研究了在高能碰撞物理中利用喷注 constituents 的动力学信息进行分类的挑战,探讨了单一性能指标可能导致的模型依赖性和偏见问题,并展示了高性能但低鲁棒性的网络的后果。

Comments 6 pages, 2 figures and 2 tables. Version presented at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: Machine Learning and the Physical Sciences. 6 December, 2025; San Diego, California, USA

详情

展开后加载摘要…

URL PDF HTML 收藏
2406.05983 2026-04-01 eess.AS

Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation

分离与重建:面向语音分离的非对称编码器-解码器

Ui-Hyeop Shin, Sangyoun Lee, Taehan Kim, Hyung-Min Park

AI总结 本文提出非对称编码器-解码器结构,通过扩展特征序列维度实现早期特征分离,结合全局与局部Transformer块提升长序列处理效率,实验表明该结构在多种基准数据集上取得state-of-the-art性能。

Comments In NeurIPS 2024; Project Page: https://dmlguq456.github.io/SepReformer_Demo

详情

展开后加载摘要…

URL PDF HTML 收藏