arXiv subjects

Zhiyuan Qi

Publications and source records attributed to Zhiyuan Qi.

13 recordsLinked to original sources

arXiv2026

RouteGraph-Mona: Confusion-Aware Routing Fine-Tuning for Mineral Image Classification

Mineral image classification is important for geological exploration and resource development, but it remains challenging due to substantial intra-class variations in appearance and high inter-class visual similarity. Multi-cognitive Visual Adapter (Mona) is a vision-oriented parameter-efficient adapter that adapts pre-trained visual models by tuning only a few parameters. However, Mona statically aggregates responses from multiple scales, limiting its ability to accommodate sample-specific scale preferences and model confusion among visually similar mineral categories. To address this issue, we propose \textbf{RouteGraph-Mona}, a lightweight route-space regularization method built on Mona. Specifically, we replace Mona's static multi-scale aggregation with sample-adaptive routing. The resulting branch-selection behavior defines a compact routing space that captures each image's scale preferences. We then regularize the resulting routing signatures with class-wise route anchors and confusion-weighted margins. The route anchors encourage class-consistent routing patterns, while the margins promote greater separation between visually similar categories in the routing space. Experiments on three public mineral image datasets with two visual backbones show that RouteGraph-Mona consistently outperforms Mona in mean accuracy and remains competitive with representative fine-tuning methods and mineral image classification baselines.

cs.CV↗

arXiv2026

Weighted Reverse Convolution for Feature Upsampling

Pre-trained vision foundation models (VFMs) provide strong semantic representations, yet their patch-level features are inherently coarse, limiting their effectiveness on tasks requiring fine-grained localization, dense prediction, and point-wise correspondence. In this work, we revisit feature upsampling for VFMs from the perspective of \textbf{\textit{inverse problem}} and propose Weighted Reverse Convolution (WRC), a spatially adaptive inverse operator for densifying high-level visual descriptors. Specifically, we formulate feature upsampling as a weighted Tikhonov-regularized least-squares problem, where spatially varying weights modulate both data fidelity and prior strength at each spatial location. This allows WRC to adapt the reconstruction to spatially varying feature characteristics, thereby preserving critical structures while mitigating over-smoothing. Moreover, WRC retains an efficient, fully differentiable closed-form FFT solution, making it a practical drop-in upsampling operator. Integrated into a lightweight self-supervised densification framework, WRC consistently improves dense feature quality across various downstream benchmarks, including segmentation, depth estimation, video object segmentation, object discovery, and keypoint correspondence, while maintaining high computational efficiency.

cs.CV↗

arXiv2026

LiveVLN: Breaking the Stop-and-Go Loop in Vision-Language Navigation

Recent navigation systems achieve strong benchmark results, yet real-world deployment often remains visibly stop-and-go. This bottleneck arises because the sense-inference-execution loop is still blocking: after each new observation, the controller must wait for sensing, transmission, and inference before motion can continue. Reducing action-generation cost alone therefore does not remove redundant waiting. To address this issue, we present LiveVLN, a training-free framework for more continuous embodied navigation by augmenting pretrained VLM navigators with multi-step action continuation. Instead of pausing for each full sense-and-inference round, LiveVLN overlaps execution with the processing of newly arrived observations, allowing refreshed future actions to be handed off before the current executable prefix is exhausted. This design keeps actions continuously available during motion, reducing idle waiting and enabling smoother online execution. The framework operates at runtime and can be integrated with compatible pretrained VLM navigators. Across R2R and RxR, LiveVLN preserves benchmark performance while reducing waiting time and improving action availability. In real-world deployments, it cuts average episode waiting time by up to $77.7\%$ and shortens wall-clock episode time by $12.6\%$ on StreamVLN and $19.6\%$ on NaVIDA, yielding more coherent execution during deployment. Code is available at https://github.com/NIneeeeeem/LiveVLN.

cs.RO↗

arXiv2025

Semantic Communication Enabled Holographic Video Processing and Transmission

Holographic video communication is considered a paradigm shift in visual communications, becoming increasingly popular for its ability to offer immersive experiences. This article provides an overview of holographic video communication and outlines the requirements of a holographic video communication system. Particularly, following a brief review of semantic com- munication, an architecture for a semantic-enabled holographic video communication system is presented. Key technologies, including semantic sampling, joint semantic-channel coding, and semantic-aware transmission, are designed based on the proposed architecture. Two related use cases are presented to demonstrate the performance gain of the proposed methods. Finally, potential research topics are discussed to pave the way for the realization of semantic-enabled holographic video communications.

eess.IV↗

arXiv2025

Adaptive Sampling and Joint Semantic-Channel Coding under Dynamic Channel Environment

Deep learning enabled semantic communications are attracting extensive attention. However, most works normally ignore the data acquisition process and suffer from robustness issues under dynamic channel environment. In this paper, we propose an adaptive joint sampling-semantic-channel coding (Adaptive-JSSCC) framework. Specifically, we propose a semantic-aware sampling and reconstruction method to optimize the number of samples dynamically for each region of the images. According to semantic significance, we optimize sampling matrices for each region of the most individually and obtain a semantic sampling ratio distribution map shared with the receiver. Through the guidance of the map, high-quality reconstruction is achieved. Meanwhile, attention-based channel adaptive module (ACAM) is designed to overcome the neural network model mismatch between the training and testing channel environment during sampling-reconstruction and encoding-decoding. To this end, signal-to-noise ratio (SNR) is employed as an extra parameter input to integrate and reorganize intermediate characteristics. Simulation results show that the proposed Adaptive-JSSCC effectively reduces the amount of data acquisition without degrading the reconstruction performance in comparison to the state-of-the-art, and it is highly adaptable and adjustable to dynamic channel environments.

eess.IV↗

arXiv2022

Combat Data Shift in Few-shot Learning with Knowledge Graph

Many few-shot learning approaches have been designed under the meta-learning framework, which learns from a variety of learning tasks and generalizes to new tasks. These meta-learning approaches achieve the expected performance in the scenario where all samples are drawn from the same distributions (i.i.d. observations). However, in real-world applications, few-shot learning paradigm often suffers from data shift, i.e., samples in different tasks, even in the same task, could be drawn from various data distributions. Most existing few-shot learning approaches are not designed with the consideration of data shift, and thus show downgraded performance when data distribution shifts. However, it is non-trivial to address the data shift problem in few-shot learning, due to the limited number of labeled samples in each task. Targeting at addressing this problem, we propose a novel metric-based meta-learning framework to extract task-specific representations and task-shared representations with the help of knowledge graph. The data shift within/between tasks can thus be combated by the combination of task-shared and task-specific representations. The proposed model is evaluated on popular benchmarks and two constructed new challenging datasets. The evaluation results demonstrate its remarkable performance.

cs.LG↗

arXiv2021

PRASEMap: A Probabilistic Reasoning and Semantic Embedding based Knowledge Graph Alignment System

Knowledge Graph (KG) alignment aims at finding equivalent entities and relations (i.e., mappings) between two KGs. The existing approaches utilize either reasoning-based or semantic embedding-based techniques, but few studies explore their combination. In this demonstration, we present PRASEMap, an unsupervised KG alignment system that iteratively computes the Mappings with both Probabilistic Reasoning (PR) And Semantic Embedding (SE) techniques. PRASEMap can support various embedding-based KG alignment approaches as the SE module, and enables easy human computer interaction that additionally provides an option for users to feed the mapping annotations back to the system for better results. The demonstration showcases these features via a stand-alone Web application with user friendly interfaces. The demo is available at https://prasemap.qizhy.com.

cs.CL↗

arXiv2021

Unsupervised Knowledge Graph Alignment by Probabilistic Reasoning and Semantic Embedding

Knowledge Graph (KG) alignment is to discover the mappings (i.e., equivalent entities, relations, and others) between two KGs. The existing methods can be divided into the embedding-based models, and the conventional reasoning and lexical matching based systems. The former compute the similarity of entities via their cross-KG embeddings, but they usually rely on an ideal supervised learning setting for good performance and lack appropriate reasoning to avoid logically wrong mappings; while the latter address the reasoning issue but are poor at utilizing the KG graph structures and the entity contexts. In this study, we aim at combining the above two solutions and thus propose an iterative framework named PRASE which is based on probabilistic reasoning and semantic embedding. It learns the KG embeddings via entity mappings from a probabilistic reasoning system named PARIS, and feeds the resultant entity mappings and embeddings back into PARIS for augmentation. The PRASE framework is compatible with different embedding-based models, and our experiments on multiple datasets have demonstrated its state-of-the-art performance.

cs.CL↗

arXiv2020

A Comprehensive Survey on Transfer Learning

Transfer learning aims at improving the performance of target learners on target domains by transferring the knowledge contained in different but related source domains. In this way, the dependence on a large number of target domain data can be reduced for constructing target learners. Due to the wide application prospects, transfer learning has become a popular and promising area in machine learning. Although there are already some valuable and impressive surveys on transfer learning, these surveys introduce approaches in a relatively isolated way and lack the recent advances in transfer learning. Due to the rapid expansion of the transfer learning area, it is both necessary and challenging to comprehensively review the relevant studies. This survey attempts to connect and systematize the existing transfer learning researches, as well as to summarize and interpret the mechanisms and the strategies of transfer learning in a comprehensive way, which may help readers have a better understanding of the current research status and ideas. Unlike previous surveys, this survey paper reviews more than forty representative transfer learning approaches, especially homogeneous transfer learning approaches, from the perspectives of data and model. The applications of transfer learning are also briefly introduced. In order to show the performance of different transfer learning models, over twenty representative transfer learning models are used for experiments. The models are performed on three different datasets, i.e., Amazon Reviews, Reuters-21578, and Office-31. And the experimental results demonstrate the importance of selecting appropriate transfer learning models for different applications in practice.

cs.LG↗

arXiv2019

Transfer Learning Toolkit: Primers and Benchmarks

The transfer learning toolkit wraps the codes of 17 transfer learning models and provides integrated interfaces, allowing users to use those models by calling a simple function. It is easy for primary researchers to use this toolkit and to choose proper models for real-world applications. The toolkit is written in Python and distributed under MIT open source license. In this paper, the current state of this toolkit is described and the necessary environment setting and usage are introduced.

cs.LG↗

arXiv2018

Experimental Studies of Two-dimensional Laminar Jet Flows in Freely Suspended Liquid Crystal Films

Two dimensional (2D) laminar jet-----a stream of fluid that projected into a surrounding medium with the flow confined in 2D-----has both theoretical and experimental significance. We carried out 2D laminar jet experiments in freely suspended liquid crystal films (FSLCFs) of nanometers thick and centimeters in size, in which individual molecules are confined to single layers thus enable film flows with two degrees of freedom. The experimental observations are found in good agreement with the classic 2D laminar jet theory of ideal cases that assume no external coupling effects, even in fact there exist strong coupling force from the ambient air. We further investigated this air coupling effect in computer simulations, with the results indicated air has little influence on the velocity maps of flow near the nozzle. This astonishing results could be intuitively understood by considering 2D incompressibility of the films. This experiment, together with a series of our previous experiments, show for a wide range of Reynolds number, FSLCFs are excellent testing beds for 2D hydrodynamics.

cond-mat.soft↗

arXiv2015

Experimental Realization of an Incompressible Newtonian Fluid in Two Dimensions

The Brownian diffusion of micron-scale inclusions in freely suspended smectic A liquid crystal films a few nanometers thick and several millimeters in diameter depends strongly on the air surrounding the film. Near atmospheric pressure, the three-dimensionally coupled film/gas system is well described by Hughes/Pailthorpe/White hydrodynamic theory but at lower pressure (p < 70 torr), the diffusion coefficient increases substantially, tending in high vacuum toward the two-dimensional limit where it is determined by film size. In the absence of air, the films are found to be a nearly ideal physical realization of a two-dimensional, incompressible Newtonian fluid.

physics.flu-dyn↗

arXiv2014

Mutual diffusion of inclusions in freely-suspended smectic liquid crystal films

We study experimentally and theoretically the hydrodynamic interaction of pairs of circular inclusions in two-dimensional, fluid smectic membranes suspended in air. By analyzing their Brownian motion, we find that the radial mutual mobilities of identical inclusions are independent of their size but that the angular coupling becomes strongly size-dependent when their radius exceeds a characteristic hydrodynamic length. The observed dependence of the mutual mobilities on inclusion size is described well for arbitrary separations by a model that generalizes the Levine/MacKintosh theory of point-force response functions and uses a boundary-element approach to calculate the mobility matrix.

cond-mat.soft↗