参考来源

第 409–432 条，共 766 条来源

409. Longformer: The Long-Document Transformer

     Beltagy et al.
     2020-04-10

     来源定位: v2 abstract and architecture description

     核查范围: Local window attention combines with selected global attention; inference limits depend on architecture.

     查阅日期: 2026-10-15

     打开原始来源 ↗
     引用本来源的文章 (2)

     • 位置编码与长上下文：旋转相位、距离偏置和有效信息使用
     • 从注意力到 IO 感知执行：Transformer 的计算结构、复杂度与可复核分析

410. LoRA: Low-Rank Adaptation of Large Language Models

     Edward J. Hu; Yelong Shen; Phillip Wallis; Zeyuan Allen-Zhu; Yuanzhi Li; Shean Wang; Lu Wang; Weizhu Chen
     2021
     打开原始来源 ↗
     引用本来源的文章 (1)

     • What Low-Rank Adaptation Can and Cannot Represent

411. LoRA: Low-Rank Adaptation of Large Language Models

     Hu et al.
     2021-06-17

     来源定位: v2; low-rank update formulation

     核查范围: A frozen matrix W receives a trainable low-rank BA update. The number of trainable values r(d_in+d_out) differs from the full matrix size.

     查阅日期: 2026-10-15

     打开原始来源 ↗
     引用本来源的文章 (2)

     • 知识蒸馏的可迁移信息：温度、教师误差与学生容量
     • 参数高效适配的低秩假设：LoRA 的优化结构、容量与合并等价性

412. Lost in the Middle: How Language Models Use Long Contexts

     Nelson F. Liu; Kevin Lin; John Hewitt; Ashwin Paranjape; Michele Bevilacqua; Fabio Petroni; Percy Liang
     2023
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Beyond the Training Window: Position, Distance, and Extrapolation in Transformers

413. Lost in the Middle: How Language Models Use Long Contexts

     Liu et al.
     2023-07-06

     来源定位: v3 abstract; multi-document QA and key-value retrieval

     核查范围: In studied models and tasks, placement of relevant information affects accuracy; this is not a universal failure rate for later models.

     查阅日期: 2026-10-15

     打开原始来源 ↗
     引用本来源的文章 (1)

     • 检索增强生成的证据工程：召回误差、引用归因与端到端可信性

414. Low latency via redundancy

     Ashish Vulimiri; Philip Brighten Godfrey; Radhika Mittal; Justine Sherry; Sylvia Ratnasamy; Scott Shenker
     2013
     打开原始来源 ↗
     引用本来源的文章 (1)

     • The Geometry of Tail Latency: Fan-Out, Queueing, and Redundant Work

415. Low-Rank Bottleneck in Multi-head Attention Models

     Srinadh Bhojanapalli; Chulhee Yun; Ankit Singh Rawat; Sashank J. Reddi; Sanjiv Kumar
     2020
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Rank Is Not a Capacity Certificate: Attention Bottlenecks Beyond the Score Matrix

416. Lower Bounds for Oblivious Data Structures

     Riko Jacob; Kasper Green Larsen; Jesper Buus Nielsen
     2018
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Hiding the Address: Oblivious RAM, Leakage, and the Cost of Access Privacy

417. Machine learning-aided engineering of hydrolases for PET depolymerization

     Hongyuan Lu; Daniel J. Diaz; Natalie J. Czarnecki; Congzhi Zhu; Wantae Kim; Raghav Shroff; et al.
     2022-04-27

     来源定位: Abstract; Extended Data Figs. 5–9

     核查范围: 底物几何、结晶与温度影响；质量损失不等于闭环回收产率。

     打开原始来源 ↗
     引用本来源的文章 (1)

     • 酶解塑料能否形成闭环：底物可及性、酶失活与单体回收账本

418. Machine Unlearning

     Lucas Bourtoule; Varun Chandrasekaran; Christopher A. Choquette-Choo; Hengrui Jia; Adelin Travers; Baiwu Zhang; David Lie; Nicolas Papernot
     2019
     打开原始来源 ↗
     引用本来源的文章 (1)

     • What an Unlearning Test Can Actually Certify

419. Magnetic control of tokamak plasmas through deep reinforcement learning

     Degrave et al.
     2022

     来源定位: TCV magnetic shape-control experiments, simulator training and device-specific control scope.

     查阅日期: 2026-10-22

     打开原始来源 ↗
     引用本来源的文章 (1)

     • 聚变等离子体的学习控制：形状跟踪、撕裂回避与闭环可靠性

420. Maintenance of datalog materialisations revisited

     Boris Motik; Yavor Nenov; Robert Piro; Ian Horrocks
     2019
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Maintaining a Fixed Point: Datalog, Retractions, and Incremental State

421. Major role of particle fragmentation in regulating biological sequestration of CO2 by the oceans

     Nathan Briggs; Giorgio Dall’Olmo; Hervé Claustre
     2020-02-14

     来源定位: Abstract; particle fragmentation results

     核查范围: 颗粒破碎影响输送；不把颗粒浓度当沉降通量。

     打开原始来源 ↗
     引用本来源的文章 (1)

     • 海洋碳泵的深度账本：颗粒输送、破碎与真正的封存时间

422. Majorana quantization and half-integer thermal quantum Hall effect in a Kitaev spin liquid

     Y. Kasahara; T. Ohnishi; Y. Mizukami; O. Tanaka; Sixiao Ma; K. Sugii; N. Kurita; H. Tanaka; J. Nasu; Y. Motome; T. Shibauchi; Y. Matsuda
     2018
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Beyond the Absence of Order: Identifying Quantum Spin Liquids amid Disorder and Imperfect Probes

423. Making Deep Neural Networks Robust to Label Noise: A Loss Correction Approach

     Giorgio Patrini; Alessandro Rozza; Aditya Krishna Menon; Richard Nock; Lizhen Qu
     2017
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Noisy Labels and the Missing Identification Assumption

424. Making snapshot isolation serializable

     Alan Fekete; Dimitrios Liarokapis; Elizabeth O'Neil; Patrick O'Neil; Dennis Shasha
     2005
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Serializable Transactions as an Invariant-Preservation Problem

425. Mamba: Linear-Time Sequence Modeling with Selective State Spaces

     ,
     2023-12-01

     来源定位: Abstract; 原文方法、理论和实验章节见作者原始论文链接

     核查范围: 一手论文；原文研究归属原作者。本文的推导示例与建议实验方案另行标注，不冒充原论文或本站实测。

     查阅日期: 2026-10-21

     打开原始来源 ↗
     引用本来源的文章 (1)

     • 选择性状态空间模型：Mamba 的递推、扫描与记忆边界

426. Mamba: Linear-Time Sequence Modeling with Selective State Spaces

     Albert Gu; Tri Dao
     2023
     打开原始来源 ↗
     引用本来源的文章 (1)

     • What a Recurrent State Can Remember: Decay, Selectivity, and Sequence Memory

427. Manifold Mixup: Better Representations by Interpolating Hidden States

     Vikas Verma; Alex Lamb; Christopher Beckham; Amir Najafi; Ioannis Mitliagkas; David Lopez-Paz; Yoshua Bengio
     2019
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Interpolation Changes the Learning Distribution

428. Marine heatwaves under global warming

     Thomas L. Frölicher; Erich M. Fischer; Nicolas Gruber
     2018
     打开原始来源 ↗
     引用本来源的文章 (1)

     • A Moving Ocean Baseline: Marine Heatwaves, Attribution and Ecological Exposure

429. Masked Autoencoders Are Scalable Vision Learners

     Kaiming He; Xinlei Chen; Saining Xie; Yanghao Li; Piotr Dollár; Ross Girshick
     2022
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Reconstruction Is Not a Certificate of Semantic Understanding

430. MaskGIT: Masked Generative Image Transformer

     Huiwen Chang; Han Zhang; Lu Jiang; Ce Liu; William T. Freeman
     2022
     打开原始来源 ↗
     引用本来源的文章 (1)

     • What a Mask Forgets: Discrete Diffusion, Joint Dependence, and Parallel Token Generation

431. Massively parallel single-cell mitochondrial DNA genotyping and chromatin profiling

     Caleb A. Lareau; Leif S. Ludwig; Christoph Muus; Satyen H. Gohil; Tongtong Zhao; Zachary Chiang; Karin Pelka; Jeffrey M. Verboon; Wendy Luo; Elena Christian; Daniel Rosebrock; Gad Getz; Genevieve M. Boland; Fei Chen; Jason D. Buenrostro; Nir Hacohen; Catherine J. Wu; Martin J. Aryee; Aviv Regev; Vijay G. Sankaran
     2020
     打开原始来源 ↗
     引用本来源的文章 (1)

     • Genomes within Cells: Drift, Selection and the Measurement of Mitochondrial Heteroplasmy

432. Mathematical discoveries from program search with large language models

     Romera-Paredes et al.
     2023

     来源定位: Program generation, evaluation function, population database and search; evaluator validity is part of the system.

     查阅日期: 2026-10-22

     打开原始来源 ↗
     引用本来源的文章 (1)

     • 从生成猜想到可检查证明：神经符号搜索中的验证器、规格与发现