Publications

Fairness Labels in Feeds: Process Disclosures, Provenance, and User Choice Among Short-Video Creators in Mainland China (Accepted; SSCI 中科院Q1 Top)

Accepted for Social Media + Society, 2026

Accepted at Social Media + Society (SSCI 中科院Q1 Top). A study of process disclosures (“fairness labels”), content provenance, and user choice among short-video creators in Mainland China. Dandan Liu (first author); Guangrui Fan (corresponding author).

Recommended citation:
Liu, D., & Fan, G. (2026). Fairness Labels in Feeds: Process Disclosures, Provenance, and User Choice Among Short-Video Creators in Mainland China. Social Media + Society. Accepted (forthcoming).

Conformal@K: Distribution-Free Top-K Miss-Risk Control for Recommendation with Overlapping-Group and Two-Stage Guarantees (Accepted; CCF A)

Accepted for ACM Transactions on Information Systems, 2026

Accepted at ACM Transactions on Information Systems (TOIS; CCF A). Distribution-free top-k miss-risk control for recommendation with overlapping-group and two-stage guarantees.

Recommended citation:
Fan, G., Liu, D., & Pan, L. (2026). Conformal@K: Distribution-Free Top-K Miss-Risk Control for Recommendation with Overlapping-Group and Two-Stage Guarantees. ACM Transactions on Information Systems. Accepted (forthcoming).

Tool, Tutor, or Crutch?: A Grounded Theory of Cognitive Scaffolding and Offloading in AI-Assisted Programming Education (SCI 中科院Q1 Top)

Published in International Journal of STEM Education, 2026

Grounded theory study of how students use AI coding assistants as tools, tutors, and cognitive crutches in programming education, with implications for scaffolding and offloading.

Recommended citation:
Liu, D., Fan, G., & Pan, L. (2026). Tool, tutor, or crutch?: A grounded theory of cognitive scaffolding and offloading in AI-assisted programming education. International Journal of STEM Education, 13, 10. https://doi.org/10.1186/s40594-025-00592-w
Download Paper

DynaKey-GNN: An efficient dynamic key-node multi-graph neural network for spatio-temporal traffic flow forecasting (SCI 中科院Q1 Top)

Published in Engineering Applications of Artificial Intelligence, 2025

A dynamic key-node multi-graph neural network for traffic flow prediction that achieves up to 12.37% higher accuracy than state-of-the-art baselines.

Recommended citation:
Fan, G., Sabri, A. Q. M., Rahman, S. S. A., & Pan, L. (2025). DynaKey-GNN: An efficient dynamic key-node multi-graph neural network for spatio-temporal traffic flow forecasting. Engineering Applications of Artificial Intelligence, 159, 111757.

Emerging Trends in Graph Neural Networks for Traffic Flow Prediction: A Survey (SCI 中科院Q1 Top)

Published in Archives of Computational Methods in Engineering, 2025

A comprehensive survey of GNN applications in traffic flow prediction from 2020 to 2024, with extensive quantitative analysis of around 40 state-of-the-art models.

Recommended citation:
Fan, G., Sabri, A. Q. M., Rahman, S. S. A., Pan, L., & Rahardja, S. (2025). Emerging Trends in Graph Neural Networks for Traffic Flow Prediction: A Survey. Archives of Computational Methods in Engineering, 1-45.

The impact of AI-assisted pair programming on student motivation, programming anxiety, collaborative learning, and programming performance: a comparative study with traditional pair programming and individual approaches (SCI 中科院Q1 Top, ESI高被引)

Published in International Journal of STEM Education, 2025

🔥 ESI Highly Cited Paper (ESI 高被引论文)
🏅 Most cited articles (past two years) in International Journal of STEM Education
📈 Most popular articles (past 9 months) in International Journal of STEM Education (Rank 1)

A quasi-experimental study with 234 undergraduate students investigating the impact of AI-assisted pair programming on motivation, anxiety, and performance.

Recommended citation:
Fan, G., Liu, D., Zhang R., & Pan, L. (2025). The impact of AI-assisted pair programming on student motivation, programming anxiety, collaborative learning, and programming performance: a comparative study with traditional pair programming and individual approaches. International Journal of STEM Education, 12(1), 16.

From Solo Post to Shared Space: How a Public LLM Agent Reshapes Human-to-Human Conversation Structure on a Social Platform

Accepted for ACM Transactions on Social Computing, 2026

Accepted at ACM Transactions on Social Computing (TSC).

Recommended citation:
Liu, D., Pan, L., Sabri, A.Q.M., & Fan, G. (2026). From Solo Post to Shared Space: How a Public LLM Agent Reshapes Human-to-Human Conversation Structure on a Social Platform. ACM Transactions on Social Computing. Accepted (forthcoming).

Skim or Swim? Investigating AI-Generated Summaries, Trust, and Comprehension in Chinese Mobile Reading (CCF B)

Published in International Journal of Human–Computer Interaction, 2025

Examining how AI-generated summaries shape engagement, comprehension, and trust among Chinese mobile readers in a mixed-methods within-subjects experiment.

Recommended citation:
Fan, G., Liu, D., & Huang, Y. (2025). Skim or Swim? Investigating AI-Generated Summaries, Trust, and Comprehension in Chinese Mobile Reading. International Journal of Human–Computer Interaction, 1–22. https://doi.org/10.1080/10447318.2025.2574513

Textless, Tiny, and Private: Hashed n-Grams with LLM-Teacher Distillation and Local Differential Privacy for Mental-Health Screening on Social Media (CCF C)

Published in IEEE Transactions on Computational Social Systems, 2026

A privacy-preserving approach for mental-health screening on social media using hashed n-grams, LLM-teacher distillation, and local differential privacy.

Recommended citation:
Liu, D., Fan, G., Guo, Q., Zhang, R., & Pan, L. (2026). Textless, Tiny, and Private: Hashed n-Grams with LLM-Teacher Distillation and Local Differential Privacy for Mental-Health Screening on Social Media. IEEE Transactions on Computational Social Systems (Early Access). https://doi.org/10.1109/TCSS.2026.3693692
Download Paper

Feeling Rules in Language Models: Mapping Norms of Emotional Appropriateness Across Roles, Institutions, and Intensity (CCF A)

Published in Annual Meeting of the Association for Computational Linguistics (ACL) 2026, 2026

This paper maps norms of emotional appropriateness in language models across roles, institutions, and intensity, revealing how feeling rules are encoded and expressed in LLMs.

Recommended citation:
Fan, G., Liu, D., Sabri, A.Q.M., Zhang, R., & Pan, L. (2026). Feeling Rules in Language Models: Mapping Norms of Emotional Appropriateness Across Roles, Institutions, and Intensity. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026). ACL.

Incomplete Multi-View Clustering via Neighborhood-Conditioned Diffusion (CCF A)

Published in International Conference on Machine Learning (ICML) 2026, 2026

Coauthored paper accepted by ICML 2026 on incomplete multi-view clustering with neighborhood-conditioned diffusion.

Recommended citation:
Guo, Q., Zuo, G., Jiang, B., Fan, G., Cui, Z., Liang, X., & Ding, J. (2026). Incomplete Multi-View Clustering via Neighborhood-Conditioned Diffusion. In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026). PMLR.

Graph-Preference Learning: Debiasing Network-Sampled Human Feedback for Target Welfare Estimation (CCF A)

Published in International Conference on Machine Learning (ICML) 2026, 2026

First-author paper accepted by ICML 2026 on debiasing network-sampled human feedback for target welfare estimation.

Recommended citation:
Fan, G., Liu, D., Sabri, A.Q.M., & Pan, L. (2026). Graph-Preference Learning: Debiasing Network-Sampled Human Feedback for Target Welfare Estimation. In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026). PMLR.

Multi-LLM Persona Generation for Virtual Focus Groups in Software Engineering: A Controlled, Multi-Domain Study of Emotional Requirements Elicitation (CCF A)

Published in ACM International Conference on the Foundations of Software Engineering (FSE) 2026, 2026

A multi-domain controlled study (mental-health, finance, fitness) comparing 4 AI configurations vs. human baselines for emotional requirements elicitation. Heterogeneous-provider plurality shows +14.7% uplift in validated requirements.

Recommended citation:
Fan, G., Liu, D., Pan, L., Zhang, R., & Guo, Q. (2026). Multi-LLM Persona Generation for Virtual Focus Groups in Software Engineering: A Controlled, Multi-Domain Study of Emotional Requirements Elicitation. In Proceedings of the ACM International Conference on the Foundations of Software Engineering (FSE 2026). ACM.

Co-Adaptive Eco-Nudging: A Privacy-Preserving Contextual Bandit with User-Taught Preferences (CCF A)

Published in CHI Conference on Human Factors in Computing Systems 2026, 2026

Two field studies with an on-device contextual bandit for personalized eco-nudging. This paper introduces the Ethical-Efficacy Frontier (EEF) and Energy ROI accounting for sustainable behavior change interventions.

Recommended citation:
Fan, G., Liu, D., et al. (2026). Co-Adaptive Eco-Nudging: A Privacy-Preserving Contextual Bandit with User-Taught Preferences. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI 2026). ACM.

When Help Hurts: Verification Load and Fatigue with AI Coding Assistants (CCF A)

Published in CHI Conference on Human Factors in Computing Systems 2026, 2026

🏆 Honourable Mention Award

A mixed-methods study comparing Inline, Chat, and Structured AI coding assistant modes. This paper introduces the “verification-load” index capturing failures, churn, and context switches to understand when AI assistance becomes counterproductive.

Recommended citation:
Fan, G., Liu, D., et al. (2026). When Help Hurts: Verification Load and Fatigue with AI Coding Assistants. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI 2026). ACM.

Is It Still You? Attributing Authorship and Authenticity in AI-Assisted Romantic Communication (CCF A)

Published in CHI Conference on Human Factors in Computing Systems 2026, 2026

This paper investigates how AI-assisted romantic communication affects perceived authorship and authenticity. Through two studies (N=152 instrumented authoring + N=704 mixed-effects experiment), we demonstrate a competence-integrity dissociation in AI-assisted romantic messages.

Recommended citation:
Fan, G., Liu, D., et al. (2026). Is It Still You? Attributing Authorship and Authenticity in AI-Assisted Romantic Communication. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI 2026). ACM.

Audit-of-Audits for the Web: Bayesian Meta-Evaluation that Yields Interval-Valued, Threshold-Aligned Fairness Claims (CCF A)

Published in The Web Conference (WWW) 2026, 2026

This paper presents a Bayesian meta-evaluation framework for auditing fairness claims on the web, yielding interval-valued, threshold-aligned fairness assessments for more robust and interpretable fairness guarantees.

Recommended citation:
Fan, G., Liu, D., Sabri, A.Q.M., & Pan, L. (2026). Audit-of-Audits for the Web: Bayesian Meta-Evaluation that Yields Interval-Valued, Threshold-Aligned Fairness Claims. In Proceedings of The Web Conference 2026 (WWW 2026).

Mind the Gap: Predicting, Explaining and Reducing Time-to-First-Comment (Reply Gap) in Online Mental-Health Communities (CCF A)

Published in AAAI Conference on Artificial Intelligence (AAAI 2026), 2026

This paper unifies predictive modeling, causal inference, and offline policy evaluation to demonstrate how an AI system can identify Reddit users in distress at risk of being ignored and proactively route effective emotional support to them faster.

Recommended citation:
Fan, G., Liu, D., & Pan, L. (2026). Mind the Gap: Predicting, Explaining and Reducing Time-to-First-Comment (Reply Gap) in Online Mental-Health Communities. In Proceedings of the 40th AAAI Conference on Artificial Intelligence (AAAI 2026).

Creative Momentum Transfer: How Timing and Labeling of AI Suggestions Shape Iterative Human Ideation (CCF A)

Published in International Joint Conferences on Artificial Intelligence (IJCAI), 2025

Investigating how the timing and labeling of AI prompts shape multi-round human ideation in a between-subjects experiment with 247 participants.

Recommended citation:
Fan, G., Liu, D., Pan, L., & Huang, Y. (2025). Creative Momentum Transfer: How Timing and Labeling of AI Suggestions Shape Iterative Human Ideation. In International Joint Conferences on Artificial Intelligence (IJCAI).

FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness (CCF B)

Accepted for Findings of the Association for Computational Linguistics: EMNLP 2026, 2026

Accepted to Findings of the Association for Computational Linguistics: EMNLP 2026.

Recommended citation:
Fan, G., Liu, D., Guo, Q., Niu, X., & Pan, L. (2026). FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness. In Findings of the Association for Computational Linguistics: EMNLP 2026. Accepted.

Consent Boundaries by Play: A CI-Grounded Micro-Game for Eliciting Stable Preferences and Diagnosing Opt-Out Friction in Web Consent (CCF B)

Published in European Conference on Computer-Supported Cooperative Work (ECSCW 2026), 2026

This paper presents a contextual-integrity-grounded micro-game for eliciting web consent boundaries, separating stable preferences, UI-induced noise, and opt-out friction across a two-session study.

Recommended citation:
Fan, G., Liu, D., Pan, L., Zhang, R., & Sabri, A.Q.M. (2026). Consent Boundaries by Play: A CI-Grounded Micro-Game for Eliciting Stable Preferences and Diagnosing Opt-Out Friction in Web Consent. In Proceedings of the 2026 European Conference on Computer-Supported Cooperative Work (ECSCW 2026). EUSSET.

LePER: Label-Free Edge Polarity Reweighting for Heterophily (CCF B)

Published in 2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2026

🎤 Oral Presentation

A novel label-free edge polarity reweighting approach for graph neural networks under heterophily settings.

Recommended citation:
Fan, G., Pan, L., Zhang, M., & Zhao, M. (2026). LePER: Label-Free Edge Polarity Reweighting for Heterophily. In Proceedings of the 2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2026). IEEE.