arXiv subjects

Kui Huang

Publications and source records attributed to Kui Huang.

12 recordsLinked to original sources

arXiv2025

PaddleOCR 3.0 Technical Report

This technical report introduces PaddleOCR 3.0, an Apache-licensed open-source toolkit for OCR and document parsing. To address the growing demand for document understanding in the era of large language models, PaddleOCR 3.0 presents three major solutions: (1) PP-OCRv5 for multilingual text recognition, (2) PP-StructureV3 for hierarchical document parsing, and (3) PP-ChatOCRv4 for key information extraction. Compared to mainstream vision-language models (VLMs), these models with fewer than 100 million parameters achieve competitive accuracy and efficiency, rivaling billion-parameter VLMs. In addition to offering a high-quality OCR model library, PaddleOCR 3.0 provides efficient tools for training, inference, and deployment, supports heterogeneous hardware acceleration, and enables developers to easily build intelligent document applications.

cs.CV↗

arXiv2025

PP-DocBee2: Improved Baselines with Efficient Data for Multimodal Document Understanding

This report introduces PP-DocBee2, an advanced version of the PP-DocBee, designed to enhance multimodal document understanding. Built on a large multimodal model architecture, PP-DocBee2 addresses the limitations of its predecessor through key technological improvements, including enhanced synthetic data quality, improved visual feature fusion strategy, and optimized inference methodologies. These enhancements yield an $11.4\%$ performance boost on internal benchmarks for Chinese business documents, and reduce inference latency by $73.0\%$ to the vanilla version. A key innovation of our work is a data quality optimization strategy for multimodal document tasks. By employing a large-scale multimodal pre-trained model to evaluate data, we apply a novel statistical criterion to filter outliers, ensuring high-quality training data. Inspired by insights into underutilized intermediate features in multimodal models, we enhance the ViT representational capacity by decomposing it into layers and applying a novel feature fusion strategy to improve complex reasoning. The source code and pre-trained model are available at \href{https://github.com/PaddlePaddle/PaddleMIX}{https://github.com/PaddlePaddle/PaddleMIX}.

cs.CV↗

arXiv2025

Qwen Look Again: Guiding Vision-Language Reasoning Models to Re-attention Visual Information

Inference time scaling drives extended reasoning to enhance the performance of Vision-Language Models (VLMs), thus forming powerful Vision-Language Reasoning Models (VLRMs). However, long reasoning dilutes visual tokens, causing visual information to receive less attention and may trigger hallucinations. Although introducing text-only reflection processes shows promise in language models, we demonstrate that it is insufficient to suppress hallucinations in VLMs. To address this issue, we introduce Qwen-LookAgain (Qwen-LA), a novel VLRM designed to mitigate hallucinations by incorporating a vision-text reflection process that guides the model to re-attention visual information during reasoning. We first propose a reinforcement learning method Balanced Reflective Policy Optimization (BRPO), which guides the model to decide when to generate vision-text reflection on its own and balance the number and length of reflections. Then, we formally prove that VLRMs lose attention to visual tokens as reasoning progresses, and demonstrate that supplementing visual information during reflection enhances visual attention. Therefore, during training and inference, Visual Token COPY and Visual Token ROUTE are introduced to force the model to re-attention visual information at the visual level, addressing the limitations of text-only reflection. Experiments on multiple visual QA datasets and hallucination metrics indicate that Qwen-LA achieves leading accuracy performance while reducing hallucinations. Our code is available at: https://github.com/Liar406/Look_Again

cs.CV↗

arXiv2025

PP-DocBee: Improving Multimodal Document Understanding Through a Bag of Tricks

With the rapid advancement of digitalization, various document images are being applied more extensively in production and daily life, and there is an increasingly urgent need for fast and accurate parsing of the content in document images. Therefore, this report presents PP-DocBee, a novel multimodal large language model designed for end-to-end document image understanding. First, we develop a data synthesis strategy tailored to document scenarios in which we build a diverse dataset to improve the model generalization. Then, we apply a few training techniques, including dynamic proportional sampling, data preprocessing, and OCR postprocessing strategies. Extensive evaluations demonstrate the superior performance of PP-DocBee, achieving state-of-the-art results on English document understanding benchmarks and even outperforming existing open source and commercial models in Chinese document understanding. The source code and pre-trained models are publicly available at \href{https://github.com/PaddlePaddle/PaddleMIX}{https://github.com/PaddlePaddle/PaddleMIX}.

cs.CV↗

arXiv2025

Unsupervised Waste Classification By Dual-Encoder Contrastive Learning and Multi-Clustering Voting (DECMCV)

Waste classification is crucial for improving processing efficiency and reducing environmental pollution. Supervised deep learning methods are commonly used for automated waste classification, but they rely heavily on large labeled datasets, which are costly and inefficient to obtain. Real-world waste data often exhibit category and style biases, such as variations in camera angles, lighting conditions, and types of waste, which can impact the model's performance and generalization ability. Therefore, constructing a bias-free dataset is essential. Manual labeling is not only costly but also inefficient. While self-supervised learning helps address data scarcity, it still depends on some labeled data and generally results in lower accuracy compared to supervised methods. Unsupervised methods show potential in certain cases but typically do not perform as well as supervised models, highlighting the need for an efficient and cost-effective unsupervised approach. This study presents a novel unsupervised method, Dual-Encoder Contrastive Learning with Multi-Clustering Voting (DECMCV). The approach involves using a pre-trained ConvNeXt model for image encoding, leveraging VisionTransformer to generate positive samples, and applying a multi-clustering voting mechanism to address data labeling and domain shift issues. Experimental results demonstrate that DECMCV achieves classification accuracies of 93.78% and 98.29% on the TrashNet and Huawei Cloud datasets, respectively, outperforming or matching supervised models. On a real-world dataset of 4,169 waste images, only 50 labeled samples were needed to accurately label thousands, improving classification accuracy by 29.85% compared to supervised models. This method effectively addresses style differences, enhances model generalization, and contributes to the advancement of automated waste classification.

cs.CV↗

arXiv2024

Digital Modeling of Massage Techniques and Reproduction by Robotic Arms

This paper explores the digital modeling and robotic reproduction of traditional Chinese medicine (TCM) massage techniques. We adopt an adaptive admittance control algorithm to optimize force and position control, ensuring safety and comfort. The paper analyzes key TCM techniques from kinematic and dynamic perspectives, and designs robotic systems to reproduce these massage techniques. The results demonstrate that the robot successfully mimics the characteristics of TCM massage, providing a foundation for integrating traditional therapy with modern robotics and expanding assistive therapy applications.

cs.RO↗

arXiv2024

RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer

In this report, we present RT-DETRv2, an improved Real-Time DEtection TRansformer (RT-DETR). RT-DETRv2 builds upon the previous state-of-the-art real-time detector, RT-DETR, and opens up a set of bag-of-freebies for flexibility and practicality, as well as optimizing the training strategy to achieve enhanced performance. To improve the flexibility, we suggest setting a distinct number of sampling points for features at different scales in the deformable attention to achieve selective multi-scale feature extraction by the decoder. To enhance practicality, we propose an optional discrete sampling operator to replace the grid_sample operator that is specific to RT-DETR compared to YOLOs. This removes the deployment constraints typically associated with DETRs. For the training strategy, we propose dynamic data augmentation and scale-adaptive hyperparameters customization to improve performance without loss of speed. Source code and pre-trained models will be available at https://github.com/lyuwenyu/RT-DETR.

cs.CV↗

arXiv2023

Towards Unified AI Drug Discovery with Multiple Knowledge Modalities

In recent years, AI models that mine intrinsic patterns from molecular structures and protein sequences have shown promise in accelerating drug discovery. However, these methods partly lag behind real-world pharmaceutical approaches of human experts that additionally grasp structured knowledge from knowledge bases and unstructured knowledge from biomedical literature. To bridge this gap, we propose KEDD, a unified, end-to-end, and multimodal deep learning framework that optimally incorporates both structured and unstructured knowledge for vast AI drug discovery tasks. The framework first extracts underlying characteristics from heterogeneous inputs, and then applies multimodal fusion for accurate prediction. To mitigate the problem of missing modalities, we leverage multi-head sparse attention and a modality masking mechanism to extract relevant information robustly. Benefiting from integrated knowledge, our framework achieves a deeper understanding of molecule entities, brings significant improvements over state-of-the-art methods on a wide range of tasks and benchmarks, and reveals its promising potential in assisting real-world drug discovery.

cs.LG↗

arXiv2022

Approaching a Minimal Topological Electronic Structure in Antiferromagnetic Topological Insulator MnBi2Te4 via Surface Modification

The topological electronic structure plays a central role in the non-trivial physical properties in topological quantum materials. A minimal, hydrogen-atom-like topological electronic structure is desired for researches. In this work, we demonstrate an effort towards the realization of such a system in the intrinsic magnetic topological insulator MnBi2Te4, by manipulating the topological surface state (TSS) via surface modification. Using high resolution laser- and synchrotron-based angle-resolved photoemission spectroscopy (ARPES), we found the TSS in MnBi2Te4 is heavily hybridized with a trivial Rashba-type surface state (RSS), which could be efficiently removed by the in situ surface potassium (K) dosing. By employing multiple experimental methods to characterize K dosed surface, we attribute such a modification to the electrochemical reactions of K clusters on the surface. Our work not only gives a clear band assignment in MnBi2Te4, but also provides possible new routes in accentuating the topological behavior in the magnetic topological quantum materials.

cond-mat.mtrl-sci↗

arXiv2022

Evidences for the exciton gas phase and its condensation in monolayer 1T-ZrTe2

The excitonic insulator (EI) is a Bose-Einstein condensation (BEC) of excitons bound by electron-hole interaction in a solid, which could support high-temperature BEC transition. The material realization of EI has been elusive, which is further challenged by the difficulty of distinguishing it from a conventional charge density wave (CDW) state. In the BEC limit, the pre-condensation exciton gas phase is a hallmark to distinguish EI from conventional CDW, yet direct experimental evidence has been lacking. Here we report a distinct correlated phase beyond the $2\times2$ CDW ground state emerging in epitaxially grown monolayer 1T-ZrTe2 and its investigation by angle-resolved photoemission spectroscopy (ARPES) and scanning tunneling microscopy (STM). The results show novel band- and energy-dependent folding behavior in a two-step process, evidenced by an exciton gas phase prior to its condensation into the final CDW state. The excellent agreement between experiments and theoretical predictions on the recovery of the pristine band structure by carrier-density-dependent suppression of the CDW state further corroborates the monolayer 1T-ZrTe2 as an EI. Our findings provide a versatile two-dimensional platform that allows tuning of the excitonic effect.

cond-mat.mtrl-sci↗

arXiv2021

Weak Antilocalization Effect up to ~ 120 K in the van der Waals Crystal Fe5-xGeTe2 with Near Room Temperature Ferromagnetism

The weak antilocalization (WAL) effect is known as a quantum correction to the classical conductivity, which never appeared in two-dimensional magnets. In this work, we reported the observation of a WAL effect in the van der Waals ferromagnet Fe5-xGeTe2 with a Curie temperature Tc ~ 270 K, which can even reach as high as ~ 120 K. The WAL effect could be well described by the Hikami-Larkin-Nagaoka and Maekawa-Fukuyama theories in the presence of strong spin-orbit coupling (SOC). Moreover, A crossover from a peak to dip behavior around 60 K in both the magnetoresistance and magnetoconductance was observed, which could be ascribed to a rare example of temperature driven Lifshitz transition as indicated by the angle-resolved photoemission spectroscopy measurements and first principles calculations. The reflective magnetic circular dichroism measurements indicate a possible spin reorientation that kills the WAL effect above 120 K. Our findings present a rare example of WAL effect in two-dimensional ferromagnet and also a magnetotransport fingerprint of the strong SOC in Fe5-xGeTe2. The results would be instructive for understanding the interaction Hamiltonian for such high Tc itinerant ferromagnetism as well as be helpful for the design of next-generation room temperature spintronic or twistronic devices.

cond-mat.mtrl-sci↗

arXiv2020

Electronic structure of a Si-containing topological Dirac semimetal CaAl2Si2

There has been an upsurge in the discovery of topological quantum materials, where various topological insulators and semimetals have been theoretically predicted and experimentally observed. However, only very few of them contains silicon, the most widely used element in electronic industry. Recently, ternary compound CaAl2Si2 has been predicted to be a topological Dirac semimetal, hosting Lorentz-symmetry-violating quasiparticles with a strongly tilted conical band dispersion. In this work, by using high-resolution angle-resolved photoemission spectroscopy (ARPES), we investigated the comprehensive electronic structure of CaAl2Si2. A pair of topological Dirac crossings is observed along the kz direction, in good agreement with the ab initio calculations, confirming the topological Dirac semimetal nature of the compound. Our study expands the topological material family on Si-containing compounds, which have great application potential in realizing low-cost, nontoxic electronic device with topological quantum states.

cond-mat.mtrl-sci↗