Hasil untuk "Art"

Guillaume Lample, Alexis Conneau

Recent studies have demonstrated the efficiency of generative pretraining for English natural language understanding. In this work, we extend this approach to multiple languages and show the effectiveness of cross-lingual pretraining. We propose two methods to learn cross-lingual language models (XLMs): one unsupervised that only relies on monolingual data, and one supervised that leverages parallel data with a new cross-lingual language model objective. We obtain state-of-the-art results on cross-lingual classification, unsupervised and supervised machine translation. On XNLI, our approach pushes the state of the art by an absolute gain of 4.9% accuracy. On unsupervised machine translation, we obtain 34.3 BLEU on WMT’16 German-English, improving the previous state of the art by more than 9 BLEU. On supervised machine translation, we obtain a new state of the art of 38.5 BLEU on WMT’16 Romanian-English, outperforming the previous best approach by more than 4 BLEU. Our code and pretrained models will be made publicly available.

Rodrigo Nogueira, Kyunghyun Cho

Recently, neural models pretrained on a language modeling task, such as ELMo (Peters et al., 2017), OpenAI GPT (Radford et al., 2018), and BERT (Devlin et al., 2018), have achieved impressive results on various natural language processing tasks such as question-answering and natural language inference. In this paper, we describe a simple re-implementation of BERT for query-based passage re-ranking. Our system is the state of the art on the TREC-CAR dataset and the top entry in the leaderboard of the MS MARCO passage retrieval task, outperforming the previous state of the art by 27% (relative) in MRR@10. The code to reproduce our results is available at this https URL

Golnaz Ghiasi, Tsung-Yi Lin, Ruoming Pang et al.

Current state-of-the-art convolutional architectures for object detection are manually designed. Here we aim to learn a better architecture of feature pyramid network for object detection. We adopt Neural Architecture Search and discover a new feature pyramid architecture in a novel scalable search space covering all cross-scale connections. The discovered architecture, named NAS-FPN, consists of a combination of top-down and bottom-up connections to fuse features across scales. NAS-FPN, combined with various backbone models in the RetinaNet framework, achieves better accuracy and latency tradeoff compared to state-of-the-art object detection models. NAS-FPN improves mobile detection accuracy by 2 AP compared to state-of-the-art SSDLite with MobileNetV2 model in [32] and achieves 48.3 AP which surpasses Mask R-CNN [10] detection accuracy with less computation time.

Chi Sun, Xipeng Qiu, Yige Xu et al.

Language model pre-training has proven to be useful in learning universal language representations. As a state-of-the-art language model pre-training model, BERT (Bidirectional Encoder Representations from Transformers) has achieved amazing results in many language understanding tasks. In this paper, we conduct exhaustive experiments to investigate different fine-tuning methods of BERT on text classification task and provide a general solution for BERT fine-tuning. Finally, the proposed solution obtains new state-of-the-art results on eight widely-studied text classification datasets.

Yuhong Li, Xiaofan Zhang, Deming Chen

We propose a network for Congested Scene Recognition called CSRNet to provide a data-driven and deep learning method that can understand highly congested scenes and perform accurate count estimation as well as present high-quality density maps. The proposed CSRNet is composed of two major components: a convolutional neural network (CNN) as the front-end for 2D feature extraction and a dilated CNN for the back-end, which uses dilated kernels to deliver larger reception fields and to replace pooling operations. CSRNet is an easy-trained model because of its pure convolutional structure. We demonstrate CSRNet on four datasets (ShanghaiTech dataset, the UCF_CC_50 dataset, the WorldEXPO'10 dataset, and the UCSD dataset) and we deliver the state-of-the-art performance. In the ShanghaiTech Part_B dataset, CSRNet achieves 47.3% lower Mean Absolute Error (MAE) than the previous state-of-the-art method. We extend the targeted applications for counting other objects, such as the vehicle in TRANCOS dataset. Results show that CSRNet significantly improves the output quality with 15.4% lower MAE than the previous state-of-the-art approach.

Hongzi Mao, R. Netravali, Mohammad Alizadeh

Chenxi Liu, Barret Zoph, Jonathon Shlens et al.

We propose a new method for learning the structure of convolutional neural networks (CNNs) that is more efficient than recent state-of-the-art methods based on reinforcement learning and evolutionary algorithms. Our approach uses a sequential model-based optimization (SMBO) strategy, in which we search for structures in order of increasing complexity, while simultaneously learning a surrogate model to guide the search through structure space. Direct comparison under the same search space shows that our method is up to 5 times more efficient than the RL method of Zoph et al. (2018) in terms of number of models evaluated, and 8 times faster in terms of total compute. The structures we discover in this way achieve state of the art classification accuracies on CIFAR-10 and ImageNet.

Cheng-Yang Fu, W. Liu, A. Ranga et al.

The main contribution of this paper is an approach for introducing additional context into state-of-the-art general object detection. To achieve this we first combine a state-of-the-art classifier (Residual-101[14]) with a fast detection framework (SSD[18]). We then augment SSD+Residual-101 with deconvolution layers to introduce additional large-scale context in object detection and improve accuracy, especially for small objects, calling our resulting system DSSD for deconvolutional single shot detector. While these two contributions are easily described at a high-level, a naive implementation does not succeed. Instead we show that carefully adding additional stages of learned transformations, specifically a module for feed-forward connections in deconvolution and a new output module, enables this new approach and forms a potential way forward for further detection research. Results are shown on both PASCAL VOC and COCO detection. Our DSSD with $513 \times 513$ input achieves 81.5% mAP on VOC2007 test, 80.0% mAP on VOC2012 test, and 33.2% mAP on COCO, outperforming a state-of-the-art method R-FCN[3] on each dataset.

F. Hutter, H. Hoos, Kevin Leyton-Brown

J. Wobbrock, Leah Findlater, Darren Gergle et al.

R. G. V. Gioi, J. Jakubowicz, J. Morel et al.

I. Witten, Radford M. Neal, J. Cleary

M. Morari, E. Zafiriou

J. W. Getzels, M. Csíkszentmihályi

Jnaneshwar Das, Christopher Filkins, Rajesh Moharana et al.

Navagunjara Reborn: The Phoenix of Odisha was built for Burning Man 2025 as both a sculpture and an experiment-a fusion of myth, craft, and computation. This paper describes the digital-physical workflow developed for the project: a pipeline that linked digital sculpting, distributed fabrication by artisans in Odisha (India), modular structural optimization in the U.S., iterative feedback through photogrammetry and digital twins, and finally, one-shot full assembly at the art site in Black Rock Desert, Nevada. The desert installation tested not just materials, but also systems of collaboration: between artisans and engineers, between myth and technology, between cultural specificity and global experimentation. We share the lessons learned in design, fabrication, and deployment and offer a framework for future interdisciplinary projects at the intersection of cultural heritage, STEAM education, and public art. In retrospect, this workflow can be read as a convergence of many knowledge systems-artisan practice, structural engineering, mythic narrative, and environmental constraint-rather than as execution of a single fixed blueprint.

Stefanie Schneider

Vision-Language Models (VLMs) transfer visual and textual data into a shared embedding space. In so doing, they enable a wide range of multimodal tasks, while also raising critical questions about the nature of machine 'understanding.' In this paper, we examine how Explainable Artificial Intelligence (XAI) methods can render the visual reasoning of a VLM - namely, CLIP - legible in art-historical contexts. To this end, we evaluate seven methods, combining zero-shot localization experiments with human interpretability studies. Our results indicate that, while these methods capture some aspects of human interpretation, their effectiveness hinges on the conceptual stability and representational availability of the examined categories.

Еремова, Е.З.

Статья посвящена исследованию творчества современного художника Императорского фарфорового завода Татьяны Васильевны Афанасьевой в контексте традиций российского фарфора. Актуальность работы обусловлена угрозой утраты уникальных художественных и технологических навыков на предприятии в условиях современного индустриального и коммерческого давления, а также недостаточной научной освещенностью профессионального творчества современных художников завода. В исследовании использованы методы сравнительного и образно-стилистического анализа произведений, интервьюирование. В статье рассмотрены произведения Татьяны Афанасьевой, созданные в разные периоды ее творческой деятельности, выявлены особенности художественного стиля, его развитие от ранних пейзажных и натюрмортных работ к более декоративным, абстрактным сериям, отражающим современные тенденции в фарфоровом искусстве. Анализ показывает, что творчество Т.В. Афанасьевой — яркое продолжение традиций, основанных на многовековом опыте и преемственности поколений мастеров, оно служит примером сочетания современных художественных приемов с классическими канонами. Автор поднимает вопрос возможного кризиса традиционного фарфорового производства на заводе, вызванного акцентом на упрощении качественных свойств продукции и потерей индивидуальности. Исследование обогащает знания о современном состоянии декоративно-прикладного искусства России, в частности о характерных особенностях петербургского фарфора, и способствует пониманию механизмов трансляции культурного наследия. Результаты работы важны для понимания путей дальнейшего развития Императорского фарфорового завода, сохранения национального достояния и популяризации традиционной культуры через современное искусство.

Zhiqing Wang, Haoxiang Fan, Shiwei Wu et al.

The integration of Generative Artificial Intelligence (GenAI) in K-6 project-based art courses presents both opportunities and challenges for enhancing creativity, engagement, and group collaboration. This study introduces a four-phase field study, involving in total two experienced K-6 art teachers and 132 students in eight offline course sessions, to investigate the usage and impact of GenAI. Specifically, based on findings in Phases 1 and 2, we developed AskArt, an interactive interface that combines DALL-E and GPT and is tailored to support elementary school students in their art projects, and deployed it in Phases 3 and 4. Our findings revealed the benefits of GenAI in providing background information, inspirations, and personalized guidance. However, challenges in query formulation for generating expected content were also observed. Moreover, students employed varied collaboration strategies, and teachers noted increased engagement alongside concerns regarding misuse and interface suitability. This study offers insights into the effective integration of GenAI in elementary education, presents AskArt as a practical tool, and provides recommendations for educators and researchers to enhance project-based learning with GenAI technologies.

Irina F. Lobasheva, Ekaterina A. Fakhrazieva

The article analyzes the art space of Kazan as one of Russia’s cultural centers through the lens of contemporary exhibition art projects initiated by museums, exhibition halls, and galleries. It addresses both the organization of significant large-scale exhibitions in the 2020s and their scientific and creative aspects, as well as their profound semantic resonance and broad social impact. The publication is accompanied by a historiographical review that focuses on key monographs, scientific articles, online reviews, and interviews related to the historical study of the city’s cultural landmarks and their role in shaping the artistic environment of Kazan. Through selected exhibition projects, the publication reveals a palette of some current collective exhibition projects, as well as exhibitions of individual artists whose art is of particular interest. As a result, these exhibitions identify the priority contemporary themes, the moods of the artists and the audience, the latest approaches to exhibition design, and the main trends and directions in the city’s art scene. It is noted that along with the permanent museum exhibitions of classical examples of visual art, the city successfully creates and develops projects by contemporary artists in various fields. It is this area, its changes and progress, that has particularly interested and attracted the attention of the authors, and as a result of the mutual collaboration between a teacher and a student, this publication has been created. A more detailed and in-depth analysis has been conducted on the following exhibitions: “Noah’s Ark” (2023), which provides a comprehensive analysis of individual works by various artists, and two exhibitions of the “Kazan Time” project. Artists of the 1990s at the Contemporary Art Gallery of the Republic of Tatarstan (2025), featuring the creative individuality of such masters as Evgeny Golubtsov and Oleg Ivanov, and “Our Avant-Garde” at the Benois Wing of the State Russian Museum in St. Petersburg (2025), focusing on the phenomenon of the popularity of the ‘fathers’ of Russian avant-garde. The article raises questions about the future development of visual arts and the role of young artists in the 21st century. The modern development of the Kazan Art School and its role in the formation of Tatarstan’s visual arts are also discussed.

P. Schwartz