arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Ziming Lin

8 papers · 2023–2026

2023: 22024: 22025: 22026: 2
2023peak 2 in 20232026

Frequent co-authors

Papers

showing 1–8 of 8 RSS

NTIRE 2025 Challenge on Cross-Domain Few-Shot Object Detection: Methods and Results

Yuqian Fu, Xingyu Qiu, Bin Ren +59 more

cs.CV · · cited 1× · arXiv ↗ · more like this ·

Cross-Domain Few-Shot Object Detection (CD-FSOD) poses significant challenges to existing object detection and few-shot detection models when applied across domains. In conjunction with NTIRE 2025, we organized the 1st CD-FSOD Challenge, aiming to advance the performance of current object detectors on entirely novel target domains with only limited labeled data. The challenge attracted 152 registered participants, received submissions from 42 teams, and concluded with 13 teams making valid final submissions. Participants approached the task from diverse perspectives, proposing novel models that achieved new state-of-the-art (SOTA) results under both open-source and closed-source settings. In this report, we present an overview of the 1st NTIRE 2025 CD-FSOD Challenge, highlighting the proposed solutions and summarizing the results submitted by the participants.

Design and Implementation of a Resource Sharing Platform based on Cloud Computing

Luqi He, Ning Wang, Yingqing Zeng +3 more

Information Systems · Journal of big data and computing. · · cited 1× · DOI ↗ · more like this ·

Entering the era of big data, while the amount of resource information is increasing exponentially, it has greatly increased the difficulty of people's resource sharing and searching. The design goal of this resource sharing platform is to gather the power of a large number of netizens to collect and integrate a large amount of resources, so that users can enrich the resource library as resource sharers while searching for resources. This system is an online multi-person resource sharing platform designed based on the Springboot and Vue2.0 framework, which implements functions such as user registration and login, data sharing, search, and comments. It also implements resource and user information management functions in the administrator module, as well as AI image audit image function based on Yolo. This system adopts cloud computing cloud database technology and cloud computing host…

Generative bootstrap processes

Ziming Lin, Fang Han

math.ST · · cited 0× · arXiv ↗ · more like this ·

We study generative bootstrap processes obtained by resampling from fitted generative distributions. We establish necessary and sufficient conditions for their conditional weak convergence to the $P$-Brownian bridge, formulated in terms of finite-dimensional conditional weak convergence and conditional asymptotic equicontinuity. We further provide general sufficient conditions and a Giné-Zinn-type characterization relating this convergence to the $P$-Donsker property. As applications, we verify these conditions for prominent classes of generative models, including triangular normalizing flows, flow matching, score-based diffusion models, and Wasserstein generative adversarial networks.

Learning Rule with Precise Spike Timing Based on Direct Feedback Alignment

NING Limiao, WANG Ziming, LIN Zhicheng, PENG Jian, TANG Huajin

Computer Vision and Pattern Recognition · DOAJ (DOAJ: Directory of Open Access Journals) · · cited 0× · DOI ↗ · more like this ·

Due to the complex spatiotemporal dynamics of spike neurons and synapses,training spike neural networks(SNNs) is relatively challenging,and there are currently no widely accepted core training algorithms and techniques.In this paper,we propose a learning rule with precise spike timing based on direct feedback alignment(PREST-DFA).Inspired by the learning algorithm called spike layer error reassignment(SLAYER),PREST-DFA uses error signals based on spike convolution differences.The output layer iteratively calculates the error values,and utilizes direct feedback alignment(DFA) to broadcast the error to hidden layer neurons,finally achieving synaptic weights update.We implement time-driven PREST-DFA,and simulation experiments demonstrate that PREST-DFA has precise spike timing learning capabilities and good biological plausibility.Based on literature search results,this is the first time…

On the consistency of bootstrap for matching estimators

Ziming Lin, Fang Han

math.ST · · cited 0× · arXiv ↗ · more like this ·

In a landmark paper, Abadie and Imbens (2008) showed that the naive bootstrap is inconsistent when applied to nearest neighbor matching estimators of the average treatment effect with a fixed number of matches. Since then, this finding has inspired numerous efforts to address the inconsistency issue, typically by employing alternative bootstrap methods. In contrast, this paper shows that the naive bootstrap is provably consistent for the original matching estimator, provided that the number of matches, $M$, diverges. The bootstrap inconsistency identified by Abadie and Imbens (2008) thus arises solely from the use of a fixed $M$.

Mono-no-aware aesthetics in Japanese art texts (Draft 3)

Ziming Lin

Cultural Studies · Arts Culture and Language · · cited 0× · DOI ↗ · more like this ·

The concept of mono-no-aware aesthetics was first introduced by the Japanese literary scholar Motoori Norinaga inhis commentary book げんじものがたりたまのおぐし . As the first of the three central Japanese aesthetic ideas,Mono no aware has profoundly influenced the expression of Japanese artistic texts. In this paper, we will look at severalJapanese fiction and film texts to explore the emotional pursuits and philosophies of life expressed by the authors andfelt by the readers and the cultural connotations behind them.

Bootstrap consistency for general double/debiased machine learning estimators

Ziming Lin, Fang Han

math.ST · · arXiv ↗ · more like this ·

Double/debiased machine learning (DML) provides a general framework for inference with high-dimensional or otherwise complex nuisance parameters by combining Neyman-orthogonal scores with cross-fitting, thereby circumventing classical Donsker-type conditions in many modern machine-learning settings. Despite its strong empirical performance, bootstrap inference for DML estimators has received little theoretical justification. This is particularly noteworthy since bootstrap methods are suggested ad used for inference on DML estimators, even though bootstrap procedures can fail for estimators that are root-$n$ consistent and asymptotically normal. This paper fills this gap by establishing bootstrap validity for DML estimators under general exchangeably weighted resampling schemes, with Efron's bootstrap as a special case. Under exactly the same conditions required for the validity of DML…

Stabilized Inverse Probability Weighting via Isotonic Calibration

Lars van der Laan, Ziming Lin, Marco Carone +1 more

stat.ME · · arXiv ↗ · more like this ·

Inverse weighting with an estimated propensity score is widely used by estimation methods in causal inference to adjust for confounding bias. However, directly inverting propensity score estimates can lead to instability, bias, and excessive variability due to large inverse weights, especially when treatment overlap is limited. In this work, we propose a post-hoc calibration algorithm for inverse propensity weights that generates well-calibrated, stabilized weights from user-supplied, cross-fitted propensity score estimates. Our approach employs a variant of isotonic regression with a loss function specifically tailored to the inverse propensity weights. Through theoretical analysis and empirical studies, we demonstrate that isotonic calibration improves the performance of doubly robust estimators of the average treatment effect.