About

Shiqiao Li took up his position in 2012 as Weedon Professor in Asian Architecture at the University of Virginia's School of Architecture, where he teaches and researches into emerging issues in contemporary Chinese cities. His academic portfolio includes history and theory courses as well as design studio instruction, under which his students have won several first prizes in international student design competitions and were nominated and shortlisted for the RIBA President’s Medal. He studied architecture at Tsinghua University in Beijing and obtained his PhD from the AA School of Architecture and Birkbeck College, University of London. Li practiced architecture in London and Hong Kong, initiating design proposals that have been published and exhibited in journals and international exhibitions. His writings have appeared in numerous architectural and cultural journals, and his books include 'Understanding the Chinese City,' 'Architecture and Modernization,' and 'Power and Virtue, Architecture and Intellectual Change in England 1650-1730.' He has served as an external examiner for PhD degrees at RMIT University and the University of New South Wales, and as an international judge for the RIBA President’s Medal for Dissertations in 2006. Li has been a keynote speaker at various universities worldwide and has lectured extensively, including at the University of Virginia, University of Sheffield, University of Tokyo, and others. Prior to his appointment at Virginia, he taught at the AA School of Architecture, the National University of Singapore, and The Chinese University of Hong Kong.

Research topics

• Artificial Intelligence
• Data Mining
• Machine Learning
• Computer Science
• Geology
• Materials science
• Data science
• Biology
• Ecology
• Mathematics
• Econometrics
• Environmental resource management
• Environmental science
• Statistics
• Psychology
• Chemistry
• Environmental planning
• Waste management
• Soil science
• Metallurgy

Selected publications

• Characterizing prognostic and immunological traits of kidney renal clear cell carcinoma via mitochondria-associated membranes and identifying DNM1L as a potential therapeutic target using machine learning

  BMC Cancer · 2026-01-22

  articleOpen access1st author

  Mitochondrial-associated membranes (MAMs) participate in cellular metabolism, calcium signaling, and cancer reprogramming, but their role in kidney renal clear cell carcinoma (KIRC) remains unclear. Clinical/transcriptomic data of KIRC and previously reported MAMs-related genes were obtained from TCGA, E-MTAB-1980, and GSE29609. After analyzing MAMs gene expression in KIRC, 101 models were generated via 10 machine learning algorithms to select the optimal MAMs-based scoring system. Survival (Kaplan-Meier) and Cox regression analyses evaluated its prognostic value; associations with immune cell infiltration, checkpoints, and drug sensitivity were explored, and key genes were validated in vitro and in vivo. Forty-two MAMs-related genes were identified, most highly expressed in KIRC tissues. A 9-gene MAMs scoring system was built using the Stepwise Cox model. High-score patients had worse prognosis, and the system was an independent prognostic factor for KIRC. High scores correlated with advanced TNM stage/grade, elevated CTLA4/PD1 (better immunotherapy response in CTLA4+/PD1+ subgroups), and sensitivity to temsirolimus/sunitinib (low scores sensitive to sorafenib). DNM1L was a key gene; its knockdown inhibited KIRC cell proliferation, invasion, and migration in vitro. In vivo experiments further confirmed that DNM1L knockdown suppressed tumor growth in BALB/c nude mice bearing KIRC xenografts. This study identified high MAMs-related gene expression in KIRC, developed a 9-gene MAMs scoring system, and validated DNM1L as a key gene, providing new insights into KIRC prognostic biomarkers and therapeutic targets.

  PublisherDOI

• Backdoor in Seconds: Unlocking Vulnerabilities in Large Pre-trained Models via Model Editing

  2025-11-08

  articleOpen accessSenior author

  Large pre-trained models have achieved notable success across a range of downstream tasks. However, recent research shows that a type of adversarial attack (i.e., backdoor attack) can manipulate the behavior of machine learning models through contaminating their training dataset, posing significant threat in the real-world application of large pre-trained model, especially for those customized models. Therefore, addressing the unique challenges for exploring vulnerability of pre-trained models is of paramount importance. Through empirical studies on the capability for performing backdoor attack in large pre-trained models (e.g., ViT), we find the following unique challenges of attacking large pre-trained models: 1) the inability to manipulate or even access large training datasets, and 2) the substantial computational resources required for training or fine-tuning these models. To address these challenges, we establish new standards for an effective and feasible backdoor attack in the context of large pre-trained models. In line with these standards, we introduce our EDT model, an Efficient, Data-free, Training-free backdoor attack method. Inspired by model editing techniques, EDT injects an editing-based lightweight codebook into the backdoor of large pre-trained models, which replaces the embedding of the poisoned image with the target image without poisoning the training dataset or training the victim model. Our experiments, conducted across various pre-trained models such as ViT, CLIP, BLIP, and stable diffusion, and on downstream tasks including image classification, image captioning, and image generation, demonstrate the effectiveness of our method. Our code is available at https://github.com/donglgcn/Editing/

  PublisherDOI

• Subspace Constraint and Contribution Estimation for Heterogeneous Federated Learning

  2025-06-10 · 2 citations

  article

  Heterogeneous Federated Learning (HFL) has received widespread attention due to its adaptability to different models and data. The HFL approach utilizing auxiliary models for knowledge transfer can further enhance flexibility. However, existing frameworks face the challenges of local overfitting and aggregation bias. To address these issues, we propose FedSCE. By restricting specific layers of the local model updates to a subspace, FedSCE reduces the degrees of freedom of the update, enhances generalization, and mitigates the risk of overfitting. The subspace is dynamically updated to ensure coverage of the latest model update trajectory. Additionally, FedSCE evaluates client contributions based on the update distance of the auxiliary model in feature space and parameter space, achieving adaptive weighted aggregation. We validate our approach in both feature-skewed and label-skewed scenarios, demonstrating that on Office10, our method exceeds the best baseline by 3.87%. The code will be available at https://github.com/AVC2-UESTC/FedSCE.git.

  PublisherDOI

• DeepFake detection in the AIGC era: A survey, benchmarks, and future perspectives

  Information Fusion · 2025-09-17 · 6 citations

  article
  PublisherDOI

• Bridging the Gap: Enhancing Gaze-Performance Link in Children with ASD through Dual-Level Visual Guidance in MR-DMT

  ArXiv.org · 2025-10-04

  preprintOpen access

  Autism Spectrum Disorder (ASD) is marked by action imitation deficits stemming from visuomotor integration impairments, posing challenges to imitation-based learning, such as dance movement therapy in mixed reality (MR-DMT). Previous gaze-guiding interventions in ASD have mainly focused on optimizing gaze in isolation, neglecting the crucial "gaze-performance link". This study investigates enhancing this link in MR-DMT for children with ASD. Initially, we experimentally confirmed the weak link: longer gaze durations didn't translate to better performance. Then, we proposed and validated a novel dual-level visual guidance system that operates on both perceptual and transformational levels: not only directing attention to task-relevant areas but also explicitly scaffolding the translation from gaze perception to performance execution. Our results demonstrate its effectiveness in boosting the gaze-performance link, laying key foundations for more precisely tailored and effective MR-DMT interventions for ASD.

  PublisherOA PDFDOI

• Dynamic Mixing Process of Dissimilar Particles in Three-Dimensional Fluidized Beds

  ACS Omega · 2025-07-23 · 1 citations

  articleOpen access

  /s, which is about 2.3 times the radial dispersion coefficient. The particle mixing at the center and top is mainly driven by a convection mechanism; meanwhile, particle mixing at the bottom and near the wall is mainly driven by a diffusion mechanism.

  PublisherDOI

• Large language models management of complex medication regimens: a case-based evaluation

  Frontiers in Pharmacology · 2025-11-24 · 1 citations

  articleOpen access

  Background: Large language models (LLMs) have shown the ability to diagnose complex medical cases, but only limited studies have evaluated the performance of LLMs in the development of evidence-based treatment plans. The purpose of this evaluation was to test four LLMs on their ability to develop safe and efficacious treatment plans on complex patients managed in the intensive care unit (ICU). Methods: Eight high-fidelity patient cases focusing on medication management were developed by critical care clinicians including history of present illness, laboratory values, vital signs, home medications, and current medications. Four LLMs [ChatGPT (GPT-3.5), ChatGPT (GPT-4), Claude-2, and Llama-2-70b] were prompted to develop an optimized medication regimen for each case. LLM generated medication regimens were then reviewed by a panel of seven critical care clinicians to assess safety and efficacy, as defined by medication errors identified and appropriate treatment for the clinical conditions. Appropriate treatment was measured by the average rate of clinician agreement to continue each medication in the regimen and compared using analysis of variance (ANOVA). Results: Clinicians identified a median of 4.1-6.9 medication errors per recommended regimen, and life-threatening medication recommendations were present in 16.3%-57.1% of the regimens, depending on LLM. Clinicians continued LLM-recommended medications at a rate of 54.6%-67.3%, with GPT-4 having the highest rate of medication continuation among all LLMs tested (p < 0.001) and the lowest rate of life-threatening medication errors (p < 0.001). Conclusion: Caution is warranted using present LLMs for medication regimens given the number of medication errors that were identified in this pilot study. However, LLMs did demonstrate potential to serve as clinical decision support for the management of complex medication regimens given the need for domain specific prompting and testing.

  PublisherOA PDFDOI

• Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety

  2025-06-10

  articleSenior author

  Domain adaptation addresses the challenge where the distribution of target inference data differs from that of the source training data. Recently, data privacy has become a significant constraint, limiting access to the source domain. To mitigate this issue, Source-Free Domain Adaptation (SFDA) methods bypass source domain data by generating source-like data or pseudo-labeling the unlabeled target domain. However, these approaches often lack theoretical grounding. In this work, we provide a theoretical analysis of the SFDA problem, focusing on the general empirical risk of the unlabeled target domain. Our analysis offers a comprehensive understanding of how representativeness, generalization, and variety contribute to controlling the upper bound of target domain empirical risk in SFDA settings. We further explore how to balance this trade-off from three perspectives: sample selection, semantic domain alignment, and a progressive learning framework. These insights inform the design of novel algorithms. Experimental results demonstrate that our proposed method achieves state-of-the-art performance on three benchmark datasets—Office-Home, DomainNet, and VisDA-C—yielding relative improvements of 3.2%, 9.1%, and 7.5%, respectively, over the representative SFDA method, SHOT.

  PublisherDOI

• Large Language Models for Causal Discovery: Current Landscape and Future Directions

  2025-09-01

  articleSenior author

  Causal discovery (CD) and Large Language Models (LLMs) have emerged as transformative fields in artificial intelligence that have evolved largely independently. While CD specializes in uncovering cause-effect relationships from data, and LLMs excel at natural language processing and generation, their integration presents unique opportunities for advancing causal understanding. This survey examines how LLMs are transforming CD across three key dimensions: direct causal extraction from text, integration of domain knowledge into statistical methods, and refinement of causal structures. We systematically analyze approaches that leverage LLMs for CD tasks, highlighting their innovative use of metadata and natural language for causal inference. Our analysis reveals both LLMs' potential to enhance traditional CD methods and their current limitations as imperfect expert systems. We identify key research gaps, outline evaluation frameworks and benchmarks for LLM-based causal discovery, and advocate future research efforts for leveraging LLMs in causality research. As the first comprehensive examination of the synergy between LLMs and CD, this work lays the groundwork for future advances in the field.

  PublisherDOI

• Simulation on instability behaviour of roadway surrounding rock under dynamic-static loading in thick coal seam

  Engineering Failure Analysis · 2025-08-29 · 8 citations

  articleCorresponding
  PublisherDOI

Recent grants

• Collaborative Research: III: Small: Physics Guided Graph Networks for Modeling Water Dynamics in Freshwater Ecosystems

  NSF · $250k · 2023–2026

• NIH Grant R15NS053442

  NIH · $212k · 2009

Frequent coauthors

• Yun Fu

  53 shared

• Zhixuan Chu

  30 shared

• Xiao‐Yuan Jing

  Guangdong University of Petrochemical Technology

  29 shared

• Mengxuan Hu

  First Affiliated Hospital of Anhui Medical University

  27 shared

• Ronghang Zhu

  25 shared

• Frank G. Zöllner

  University Medical Centre Mannheim

  25 shared

• Haihu Liu

  25 shared

• Zhengliang Liu

  Augusta University

  24 shared

Education

• PhD, Electrical and Computer Engineering

  Northeastern University

  2017

Awards & honors

• RIBA President’s Medal (shortlisted)