Zixuan Guo

Dr. Zixuan Guo

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 September 2025 in Heritage
Verified
Journal Article
Open access
PDF file
ZGZixuan Guo
Zixuan Guo
SKSameer Kumar
Sameer Kumar
HWHoubin Wang
Houbin Wang
JLJingyi Li
Jingyi Li
This paper introduces DKR-YOLO, an advanced deep learning framework designed to empower the digital preservation and sustainable management of ancient mural heritage. Building upon YOLOv8, DKR-YOLO integrates innovative components—including the DySnake Conv layer for refined feature extraction and an Adaptive Convolutional Kernel Warehouse to optimize representation—addressing challenges posed by intricate details, diverse artistic styles, and mural degradation. The network’s architecture further incorporates a Residual Feature Augmentation (RFA)-enhanced FPN (RE-FPN), prioritizing the most critical visual features and enhancing interpretability. Extensive experiments on mural datasets demonstrate that DKR-YOLO achieves a 43.6% reduction in FLOPs, a 3.7% increase in accuracy, and a 5.1% improvement in mAP compared to baseline models. This performance, combined with an emphasis on robustness and interpretability, supports more inclusive and accessible applications of AI for cultural institutions, thereby fostering broader participation and equity in digital heritage preservation.

4 August 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
ZGZixuan Guo
Zixuan Guo
SKSameer Kumar
Sameer Kumar
This paper introduces DKR-YOLO, an advanced deep learning framework designed to empower the digital preservation and sustainable management of ancient mural heritage. Building upon YOLOv8, DKR-YOLO integrates innovative components—including the DySnake Conv layer for refined feature extraction and an Adaptive Convolutional Kernel Warehouse to optimize representation—addressing challenges posed by intricate details, diverse artistic styles, and mural degradation. The network’s architecture further incorporates a Residual Feature Augmentation (RFA)-enhanced FPN (RE-FPN) and Efficient Channel Attention (ECA), prioritizing the most critical visual features and enhancing interpretability. Extensive experiments on mural datasets demonstrate that DKR-YOLO achieves a 43.6% reduction in FLOPs, a 3.7% increase in accuracy, and a 5.1% improvement in mAP compared to baseline models. This performance, combined with an emphasis on robustness and interpretability, supports more inclusive and accessible applications of AI for cultural institutions—including small museums and local heritage organizations—thereby fostering broader participation and equity in digital heritage preservation.

11 November 2025 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
WNWenyu Ning
Wenyu Ning
SSShaotao Shi
Shaotao Shi
YYYuchen Yin
Yuchen Yin
ZGZixuan Guo
Zixuan Guo
JXJinrong Xie
Jinrong Xie
Personalized summarization seeks to tailor the summary content to an individual user’s goals, preferences, or learning needs, offering a more adaptive alternative to generic, one-size-fits-all summaries. However, current summarization systems largely neglect explicit modeling of user intent, resulting in limited alignment between generated summaries and actual user expectations. This issue is especially critical in educational contexts, where learners may differ significantly in learning objectives, prior knowledge, or preferred granularity of information. In this paper, we propose a novel framework for intent-aware personalized summarization, designed for bilingual educational settings (English and Chinese), powered by moderately sized Large Language Models (LLMs) in the 7–9 billion parameter range. We formulate personalized summarization as the task of generating a factually consistent and concise summary of an input document, guided by an explicit, structured user intent representation. To address this challenge, we introduce a hybrid architecture that integrates a user intent encoder with a pointer-generator Transformer built atop LLM backbones. Our system features a two-stage generation strategy: (1) a content selector module retrieves segments aligned with user intent; (2) a guided decoder generates coherent summaries conditioned on both content and intent embeddings. We evaluate our method on two publicly available educational datasets (EduLect-Sum for English, EduCN-Sum for Chinese), and compare against 10+ strong baselines, including both traditional (e.g., T5, BART, Pegasus) and modern LLM-based summarizers. Empirical results show that our approach consistently outperforms baselines across four standard metrics (ROUGE, BLEU, BERTScore, FactCC) and human evaluations, achieving 6–8 point ROUGE gains, 5+% BERTScore and factuality improvements, and higher user satisfaction. Further ablation studies and cross-lingual generalization experiments validate the efficacy of explicit intent modeling. Our findings highlight the value of user-intent conditioning in educational summarization and demonstrate the feasibility of building compact, bilingual, and user-centric summarization systems with LLMs.

1 July 2026 in Sage Open
Verified
Journal Article
Open access
ZGZixuan Guo
Zixuan Guo
YWYinan Wang
Yinan Wang
HWHoubin Wang
Houbin Wang
JLJingyi Li
Jingyi Li
JZJia Zhang
Jia Zhang
SKSameer Kumar
Sameer Kumar
As AI-generated content (AIGC) apps become more important in work, school, and everyday life, it’s more crucial than ever to know what keeps people using them. Users often have problems because of task-technology misalignment, forming habits, and cultural norms that have a strong foundation in society, especially in places that are changing quickly and have very different cultures. It is informed by dual-process theory, investigates the interaction within deliberate cognitive evaluations, specifically perceived usefulness (PU) and perceived ease of use (PEOU), and automatic affective responses, such as satisfaction and habitual behaviors, in influencing continuance intention, with Confucian values acting as a cultural moderator. The partial least squares structural equation modeling technique (PLS-SEM) was used to look at data from 421 Chinese AIGC users who worked in professional design locations in Beijing. The findings reveal that both task-technology fit (TTF) and individual-technology fit (ITF) exert positive influences on cognitive and affective evaluations, with ITF exhibiting a consistently more robust effect. Satisfaction fosters habit formation, and cognitive, affective, and habitual factors collectively forecast continuance intention. Confucian values lessen the effects of PU on the desire to continue and of satisfaction on the formation of habits. These results combine dual-process mechanisms, technology fit theory, and cultural moderation to give us theoretically sound ideas for keeping people interested and designing AI-driven content systems.

1 October 2026 in Information Processing & Management
Verified
Journal Article
Closed access
XWXiaoyu Wu
Xiaoyu Wu
SLShuo Li
Shuo Li
SLShuang Liu
Shuang Liu
ZGZixuan Guo
Zixuan Guo