publications

Published

2026

1. Embedding in recommender systems: A survey

   Maolin Wang, Xinjian Zhao, Wanyu Wang, and 9 more authors

   ACM Transactions on Information Systems, 2026

   Abs Bib

   Recommender systems have become an essential component of many online platforms, providing personalized recommendations to users. A crucial aspect is embedding techniques that convert the high-dimensional discrete features, such as user and item IDs, into low-dimensional continuous vectors, which can enhance the recommendation performance. Embedding techniques have revolutionized the capture of complex entity relationships, generating significant research interest. This survey presents a comprehensive analysis of recent advances in recommender system embedding techniques. We examine centralized embedding approaches across matrix, sequential, and graph structures. In matrix-based scenarios, collaborative filtering generates embeddings that effectively model user-item preferences, particularly in sparse data environments. For sequential data, we explore various approaches including …

   @article{wang2026embedding,
     pubtype = {published},
     title = {Embedding in recommender systems: A survey},
     author = {Wang, Maolin and Zhao, Xinjian and Wang, Wanyu and Zhang, Sheng and Li, Jiansheng and Yu, Bowen and Wang, Binhao and Zhou, Shucheng and Yin, Dawei and Li, Qing and Guo, Ruocheng and Zhao, Xiangyu},
     journal = {ACM Transactions on Information Systems},
     volume = {44},
     number = {5},
     pages = {1-55},
     year = {2026},
     publisher = {ACM},
     url = {https://dl.acm.org/doi/abs/10.1145/3812652},
   }

2. Report on the Workshop on Benchmarking Causal Models (CausalBench) 2026

   K. Selçuk Candan, Ruocheng Guo, Huan Liu, and 1 more author

   In WSDM, 2026

   DOI Bib

   @inproceedings{candan2026report,
     pubtype = {published},
     title = {Report on the Workshop on Benchmarking Causal Models (CausalBench) 2026},
     author = {Candan, K. Selçuk and Guo, Ruocheng and Liu, Huan and Sheth, Paras},
     booktitle = {WSDM},
     pages = {113--114},
     year = {2026},
     doi = {10.1145/3779211.3807940}
   }

3. RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning

   Xiang Gao, Yuguang Yao, Qi Zhang, and 5 more authors

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

   DOI Bib

   @inproceedings{gao2026rimrule,
     pubtype = {published},
     title = {RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning},
     author = {Gao, Xiang and Yao, Yuguang and Zhang, Qi and Dong, Kaiwen and Baidya, Avinash and Guo, Ruocheng and Hasson, Hilaf and Das, Kamalika},
     booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL)},
     pages = {34631--34646},
     year = {2026},
     doi = {10.18653/v1/2026.acl-long.1599}
   }

4. ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents

   Dawei Li, Yuguang Yao, Zhen Tan, and 2 more authors

   In ACL Findings, 2026

   DOI Bib

   @inproceedings{li2026toolprmbench,
     pubtype = {published},
     title = {ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents},
     author = {Li, Dawei and Yao, Yuguang and Tan, Zhen and Liu, Huan and Guo, Ruocheng},
     booktitle = {ACL Findings},
     pages = {12378--12391},
     year = {2026},
     doi = {10.18653/v1/2026.findings-acl.602}
   }

5. Renormalization Group Guided Tensor Network Structure Search

   Maolin Wang, Bowen Yu, Sheng Zhang, and 8 more authors

   In Proceedings of the AAAI Conference on Artificial Intelligence, 2026

   DOI Bib

   @inproceedings{wang2026renormalization,
     pubtype = {published},
     title = {Renormalization Group Guided Tensor Network Structure Search},
     author = {Wang, Maolin and Yu, Bowen and Zhang, Sheng and Mi, Linjie and Wang, Wanyu and Wang, Yiqi and Jia, Pengyue and Wei, Xuetao and Xu, Zenglin and Guo, Ruocheng and Zhao, Xiangyu},
     booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
     volume = {40},
     number = {31},
     pages = {26346--26354},
     year = {2026},
     doi = {10.1609/aaai.v40i31.39840}
   }

6. SAGE: Global Semantic Alignment with LLMs for Long-Tail Sequential Recommendation

   Maolin Wang, Tongshu Bian, Z K Wang, and 6 more authors

   In ACM Web Conference (WWW), 2026

   DOI Bib

   @inproceedings{wang2026sage,
     pubtype = {published},
     title = {SAGE: Global Semantic Alignment with LLMs for Long-Tail Sequential Recommendation},
     author = {Wang, Maolin and Bian, Tongshu and Wang, Z K and Jiang, Xiaotong and Wang, B. L. and Xu, D and Wang, Wanyu and Guo, Ruocheng and Zhao, Xiangyu},
     booktitle = {ACM Web Conference (WWW)},
     pages = {6433--6444},
     year = {2026},
     doi = {10.1145/3774904.3792456}
   }

7. Node-Level Uncertainty Estimation in LLM-Generated SQL

   Hilaf Hasson and Ruocheng Guo

   In Language Resources and Evaluation Conference (LREC), 2026

   DOI Bib

   @inproceedings{hasson2026node,
     pubtype = {published},
     title = {Node-Level Uncertainty Estimation in LLM-Generated SQL},
     author = {Hasson, Hilaf and Guo, Ruocheng},
     booktitle = {Language Resources and Evaluation Conference (LREC)},
     pages = {2145--2153},
     year = {2026},
     doi = {10.63317/3i2i8xy5e7hs}
   }

8. SONATA: Synergistic Coreset Informed Adaptive Temporal Tensor Factorization

   Maolin Wang, Zhiqi Li, Binhao Wang, and 7 more authors

   In ICLR, 2026

   Bib

   @inproceedings{wang2026sonata,
     pubtype = {published},
     title = {SONATA: Synergistic Coreset Informed Adaptive Temporal Tensor Factorization},
     author = {Wang, Maolin and Li, Zhiqi and Wang, Binhao and Chen, Xuhui and Wei, Tianshuo and Wang, Wanyu and Fang, Shikai and Guo, Ruocheng and Xu, Zenglin and Zhao, Xiangyu},
     booktitle = {ICLR},
     year = {2026}
   }

2025

1. Function Calling in Large Language Models: Industrial Practices, Challenges, and Future Directions

   Maolin Wang, Yingyi Zhang, Cunyin Peng, and 8 more authors

   2025

   Bib HTML

   @article{wang2025function,
     pubtype = {published},
     title = {Function Calling in Large Language Models: Industrial Practices, Challenges, and Future Directions},
     author = {Wang, Maolin and Zhang, Yingyi and Peng, Cunyin and Chen, Yicheng and Zhou, Wei and Gu, Jinjie and Zhuang, Chenyi and Guo, Ruocheng and Yu, Bowen and Wang, Wanyu and others},
     year = {2025},
   }

2. FLUID-MMRec: Stein-Guided Entropic Flow for Multi-Modal Sequential Recommendation

   Maolin Wang, Yutian Xiao, Binhao Wang, and 6 more authors

   In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2025

   Bib

   @inproceedings{wang2025fluid,
     pubtype = {published},
     title = {FLUID-MMRec: Stein-Guided Entropic Flow for Multi-Modal Sequential Recommendation},
     author = {Wang, Maolin and Xiao, Yutian and Wang, Binhao and Zhang, Sheng and Ye, Shanshan and Wang, Wanyu and Yin, Hongzhi and Guo, Ruocheng and Xu, Zenglin},
     booktitle = {Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
     year = {2025},
     publisher = {Association for Computing Machinery},
     address = {Toronto, Canada}
   }

3. Optimal Policy Adaptation under Covariate Shift

   Xueqing Liu, Qinwei Yang, Zhaoqing Tian, and 2 more authors

   In Proceedings of the 34th International Joint Conference on Artificial Intelligence, 2025

   Bib

   @inproceedings{liu2025optimal,
     pubtype = {published},
     title = {Optimal Policy Adaptation under Covariate Shift},
     author = {Liu, Xueqing and Yang, Qinwei and Tian, Zhaoqing and Guo, Ruocheng and Wu, Peng},
     booktitle = {Proceedings of the 34th International Joint Conference on Artificial Intelligence},
     year = {2025},
     pages = {to appear}
   }

4. DANCE: Resource-Efficient Neural Architecture Search with Data-Aware and Continuous Adaptation

   Maolin Wang, Tianshuo Wei, Sheng Zhang, and 6 more authors

   In Proceedings of the 34th International Joint Conference on Artificial Intelligence (IJCAI), 2025

   Bib

   @inproceedings{wang2025dance,
     pubtype = {published},
     title = {DANCE: Resource-Efficient Neural Architecture Search with Data-Aware and Continuous Adaptation},
     author = {Wang, Maolin and Wei, Tianshuo and Zhang, Sheng and Guo, Ruocheng and Wang, Wangyu and Ye, Shanshan and Zou, Lixin and Wei, Xuetao and Zhao, Xiangyu},
     booktitle = {Proceedings of the 34th International Joint Conference on Artificial Intelligence (IJCAI)},
     year = {2025},
     publisher = {International Joint Conferences on Artificial Intelligence Organization},
     address = {Montreal, Canada}
   }

5. STAR-Rec: Making Peace with Length Variance and Pattern Diversity in Sequential Recommendation

   Maolin Wang, Sheng Zhang, Ruocheng Guo, and 6 more authors

   In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2025

   DOI Bib

   @inproceedings{wang2025starrec,
     pubtype = {published},
     title = {STAR-Rec: Making Peace with Length Variance and Pattern Diversity in Sequential Recommendation},
     author = {Wang, Maolin and Zhang, Sheng and Guo, Ruocheng and Wang, Wanyu and Wei, Xuetao and Liu, Zitao and Yin, Hongzhi and Chang, Yi and Zhao, Xiangyu},
     booktitle = {Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval},
     year = {2025},
     pages = {to appear},
     doi = {10.1145/3726302.3730087}
   }

6. Stepwise reasoning disruption attack of llms

   Jingyu Peng, Maolin Wang, Xiangyu Zhao, and 6 more authors

   In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2025

   Abs Bib

   Large language models (LLMs) have made remarkable strides in complex reasoning tasks, but their safety and robustness in reasoning processes remain unexplored, particularly in third-party platforms that facilitate user interactions via APIs. Existing attacks on LLM reasoning are constrained by specific settings or lack of imperceptibility, limiting their feasibility and generalizability. To address these challenges, we propose the Stepwise rEasoning Error Disruption (SEED) attack, which subtly injects errors into prior reasoning steps to mislead the model into producing incorrect subsequent reasoning and final answers. Unlike previous methods, SEED is compatible with zero-shot and few-shot settings, maintains the natural reasoning flow, and ensures covert execution without modifying the instruction. Extensive experiments on four datasets across four different models demonstrate SEED’s effectiveness, revealing the vulnerabilities of LLMs to disruptions in reasoning processes. These findings underscore the need for greater attention to the robustness of LLM reasoning to ensure safety in practical applications. Our code is available at: https://github. com/Applied-Machine-Learning-Lab/SEED-Attack

   @inproceedings{peng2025stepwise,
     pubtype = {published},
     title = {Stepwise reasoning disruption attack of llms},
     author = {Peng, Jingyu and Wang, Maolin and Zhao, Xiangyu and Zhang, Kai and Wang, Wanyu and Jia, Pengyue and Liu, Qidong and Guo, Ruocheng and Liu, Qi},
     booktitle = {Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
     pages = {5040-5058},
     year = {2025},
     url = {https://aclanthology.org/2025.acl-long.251/},
   }

7. SPARK: Adaptive Low-Rank Knowledge Graph Modeling in Hybrid Geometric Spaces for Recommendation

   Binhao Wang, Yutian Xiao, Maolin Wang, and 4 more authors

   In ACM International Conference on Information and Knowledge Management (CIKM), 2025

   DOI arXiv Bib

   @inproceedings{wang2025spark,
     pubtype = {published},
     title = {SPARK: Adaptive Low-Rank Knowledge Graph Modeling in Hybrid Geometric Spaces for Recommendation},
     author = {Wang, Binhao and Xiao, Yutian and Wang, Maolin and Li, Zekang and Wei, Tianshuo and Guo, Ruocheng and Zhao, Xiangyu},
     booktitle = {ACM International Conference on Information and Knowledge Management (CIKM)},
     pages = {2956--2966},
     year = {2025},
     doi = {10.1145/3746252.3760999},
   }

8. Enhancing clinical documentation with voice processing and large language models: a study on the LAOS system

   Yupeng Xu, Huixun Jia, Maolin Wang, and 14 more authors

   NPJ Digital Medicine, 2025

   Abs Bib

   The growing volume of Electronic Health Records (EHRs) has enhanced patient care quality but significantly increased the cognitive workload on clinicians, particularly in ophthalmology where specialists handle 1.6 times more patient consultations than other specialties. This study introduces the “LLM-based Auxiliary Ophthalmic System (LAOS),” an integrated framework leveraging Large Language Models (LLMs) and audio processing to improve clinical documentation accuracy and efficiency. LAOS combines voice recognition with Retrieval-Augmented Generation (RAG) and Low-Rank Adaptation (LoRA) to convert clinical conversations into structured documentation while dynamically retrieving relevant medical knowledge. The system was evaluated across three critical documentation tasks: Admission Reports, Surgery Records, and Discharge Summaries. Through both quantitative metrics (BLEU, ROUGE-L …

   @article{xu2025enhancing,
     pubtype = {published},
     title = {Enhancing clinical documentation with voice processing and large language models: a study on the LAOS system},
     author = {Xu, Yupeng and Jia, Huixun and Wang, Maolin and Feng, Jie and Xu, Xun and Wang, Haiyan and Chen, Jieqiong and Zheng, Zheng and Yang, Xiaoyan and Shen, Yue and Wang, Jian and Zhuang, Chenyi and Wei, Peng and Guo, Ruocheng and Zhao, Xiangyu and Fan, Junxiang and Sun, Xiaodong},
     journal = {NPJ Digital Medicine},
     volume = {8},
     number = {1},
     pages = {798},
     year = {2025},
     publisher = {Nature Publishing Group UK},
     url = {https://www.nature.com/articles/s41746-025-02170-4},
   }

9. Metalora: Tensor-enhanced adaptive low-rank fine-tuning

   Maolin Wang, Xiangyu Zhao, Ruocheng Guo, and 1 more author

   In 2025 IEEE 41st International Conference on Data Engineering (ICDE), 2025

   Abs Bib

   There has been a significant increase in the deployment of neural network models, presenting substantial challenges in model adaptation and fine-tuning. Efficient adaptation is crucial in maintaining model performance across diverse tasks and domains. While Low-Rank Adaptation (LoRA) has emerged as a promising parameter-efficient fine-tuning method, its fixed parameter nature limits its ability to handle dynamic task requirements effectively. Adapting models to new tasks can be challenging due to the need for extensive fine-tuning. Current LoRA variants primarily focus on general parameter reduction while overlooking the importance of dynamic parameter adjustment and meta-learning capabilities. Moreover, existing approaches mainly address static adaptations, neglecting the potential benefits of task-aware parameter generation in handling diverse task distributions. To address these limitations, this Ph.D …

   @inproceedings{wang2025metalora,
     pubtype = {published},
     title = {Metalora: Tensor-enhanced adaptive low-rank fine-tuning},
     author = {Wang, Maolin and Zhao, Xiangyu and Guo, Ruocheng and Wang, Junhui},
     booktitle = {2025 IEEE 41st International Conference on Data Engineering (ICDE)},
     pages = {4680-4684},
     year = {2025},
     publisher = {IEEE},
     url = {https://ieeexplore.ieee.org/abstract/document/11113105/},
   }

2024

1. Confidence-aware fine-tuning of sequential recommendation systems via conformal prediction

   Chen Wang, Fangxin Wang, Ruocheng Guo, and 2 more authors

   IEEE International Conference on Big Data (BigData), 2024

   arXiv Bib

   @article{wang2024confidence,
     pubtype = {published},
     title = {Confidence-aware fine-tuning of sequential recommendation systems via conformal prediction},
     author = {Wang, Chen and Wang, Fangxin and Guo, Ruocheng and Liang, Yueqing and Yu, Philip S},
     journal = {IEEE International Conference on Big Data (BigData)},
     year = {2024}
   }

2. Conformal counterfactual inference under hidden confounding

   Zonghao Chen^†, Ruocheng Guo^†, Jean-François Ton, and 1 more author

   In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024

   Bib

   @inproceedings{chen2024conformal,
     pubtype = {published},
     title = {Conformal counterfactual inference under hidden confounding},
     author = {Chen, Zonghao and Guo, Ruocheng and Ton, Jean-Fran{\c{c}}ois and Liu, Yang},
     booktitle = {Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
     pages = {397--408},
     year = {2024}
   }

3. Learning the Optimal Policy for Balancing Short-Term and Long-Term Rewards

   Qinwei Yang, Xueqing Liu, Yan Zeng, and 3 more authors

   In Advances in Neural Information Processing Systems, 2024

   Bib

   @inproceedings{yang2024learning,
     pubtype = {published},
     title = {Learning the Optimal Policy for Balancing Short-Term and Long-Term Rewards},
     author = {Yang, Qinwei and Liu, Xueqing and Zeng, Yan and Guo, Ruocheng and Liu, Yang and Wu, Peng},
     booktitle = {Advances in Neural Information Processing Systems},
     year = {2024}
   }

4. Fair Classifiers that Abstain without Harm

   Tongxin Yin, Jean-François Ton, Ruocheng Guo, and 3 more authors

   In Proceedings of the 12th International Conference on Learning Representations, 2024

   Bib

   @inproceedings{yin2024fair,
     pubtype = {published},
     title = {Fair Classifiers that Abstain without Harm},
     author = {Yin, Tongxin and Ton, Jean-François and Guo, Ruocheng and Yao, Yuanshun and Liu, Mingyan and Liu, Yang},
     booktitle = {Proceedings of the 12th International Conference on Learning Representations},
     year = {2024}
   }

5. Large Multimodal Model Compression via Iterative Efficient Pruning and Distillation

   Maolin Wang, Yao Zhao, Jiajia Liu, and 5 more authors

   In Companion Proceedings of the ACM Web Conference 2024, 2024

   DOI Bib

   @inproceedings{wang2024large,
     pubtype = {published},
     title = {Large Multimodal Model Compression via Iterative Efficient Pruning and Distillation},
     author = {Wang, Maolin and Zhao, Yao and Liu, Jiajia and Chen, Jingdong and Zhuang, Chenyi and Gu, Jinjie and Guo, Ruocheng and Zhao, Xiangyu},
     booktitle = {Companion Proceedings of the ACM Web Conference 2024},
     pages = {235--244},
     year = {2024},
     publisher = {Association for Computing Machinery},
     doi = {10.1145/3589335.3648321}
   }

6. Tensorized hypergraph neural networks

   Maolin Wang, Yaoming Zhen, Yu Pan, and 5 more authors

   In Proceedings of the 2024 SIAM International Conference on Data Mining (SDM), 2024

   Bib

   @inproceedings{wang2024tensorized,
     pubtype = {published},
     title = {Tensorized hypergraph neural networks},
     author = {Wang, Maolin and Zhen, Yaoming and Pan, Yu and Zhao, Yao and Zhuang, Chenyi and Xu, Zenglin and Guo, Ruocheng and Zhao, Xiangyu},
     booktitle = {Proceedings of the 2024 SIAM International Conference on Data Mining (SDM)},
     pages = {127--135},
     year = {2024},
     organization = {SIAM}
   }

7. DNS-Rec: Data-aware Neural Architecture Search for Recommender Systems

   Sheng Zhang, Maolin Wang, Xiangyu Zhao, and 6 more authors

   In Proceedings of the 18th ACM Conference on Recommender Systems, 2024

   Bib

   @inproceedings{zhang2024dnsrec,
     pubtype = {published},
     title = {DNS-Rec: Data-aware Neural Architecture Search for Recommender Systems},
     author = {Zhang, Sheng and Wang, Maolin and Zhao, Xiangyu and Guo, Ruocheng and Zhao, Yao and Zhuang, Chenyi and Gu, Jinjie and Zhang, Zijian and Yin, Hongzhi},
     booktitle = {Proceedings of the 18th ACM Conference on Recommender Systems},
     year = {2024},
     pages = {591--600}
   }

8. SMLP4Rec: An Efficient All-MLP Architecture for Sequential Recommendations

   Jingtong Gao, Xiangyu Zhao, Muyang Li, and 5 more authors

   ACM Transactions on Information Systems, 2024

   Bib

   @article{gao2024smlp4rec,
     pubtype = {published},
     title = {SMLP4Rec: An Efficient All-MLP Architecture for Sequential Recommendations},
     author = {Gao, Jingtong and Zhao, Xiangyu and Li, Muyang and Zhao, Minghao and Wu, Runze and Guo, Ruocheng and Liu, Yiding and Yin, Dawei},
     journal = {ACM Transactions on Information Systems},
     volume = {42},
     number = {3},
     pages = {86:1--86:23},
     year = {2024}
   }

9. Media Bias Matters: Understanding the Impact of Politically Biased News on Vaccine Attitudes in Social Media

   Bohan Jiang, Lu Cheng, Zhen Tan, and 2 more authors

   In IEEE International Conference on Data Science and Advanced Analytics (DSAA), 2024

   DOI Bib

   @inproceedings{jiang2024media,
     pubtype = {published},
     title = {Media Bias Matters: Understanding the Impact of Politically Biased News on Vaccine Attitudes in Social Media},
     author = {Jiang, Bohan and Cheng, Lu and Tan, Zhen and Guo, Ruocheng and Liu, Huan},
     booktitle = {IEEE International Conference on Data Science and Advanced Analytics (DSAA)},
     pages = {1--10},
     year = {2024},
     doi = {10.1109/dsaa61799.2024.10722800}
   }

2023

1. AutoMLP: Automated MLP for Sequential Recommendations

   Muyang Li, Zijian Zhang, Xiangyu Zhao, and 4 more authors

   In Proceedings of the ACM Web Conference 2023, 2023

   Abs Bib

   Sequential recommender systems aim to predict users’ next interested item given their historical interactions. However, a long-standing issue is how to distinguish between users’ long/short-term interests, which may be heterogeneous and contribute differently to the next recommendation. Existing approaches usually set pre-defined short-term interest length by exhaustive search or empirical experience, which is either highly inefficient or yields subpar results. The recent advanced transformer-based models can achieve state-of-the-art performances despite the aforementioned issue, but they have a quadratic computational complexity to the length of the input sequence. To this end, this paper proposes a novel sequential recommender system, AutoMLP, aiming for better modeling users’ long/short-term interests from their historical interactions. In addition, we design an automated and adaptive search algorithm for …

   @inproceedings{li2023automlp,
     pubtype = {published},
     title = {AutoMLP: Automated MLP for Sequential Recommendations},
     author = {Li, Muyang and Zhang, Zijian and Zhao, Xiangyu and Wang, Wanyu and Zhao, Minghao and Wu, Runze and Guo, Ruocheng},
     booktitle = {Proceedings of the ACM Web Conference 2023},
     year = {2023},
     url = {https://dl.acm.org/doi/abs/10.1145/3543507.3583440},
   }

2. Virtual Node Tuning for Few-shot Node Classification

   Zhen Tan, Ruocheng Guo, Kaize Ding, and 1 more author

   In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2023

   Abs Bib

   Few-shot Node Classification (FSNC) is a challenge in graph representation learning where only a few labeled nodes per class are available for training. To tackle this issue, meta-learning has been proposed to transfer structural knowledge from base classes with abundant labels to target novel classes. However, existing solutions become ineffective or inapplicable when base classes have no or limited labeled nodes. To address this challenge, we propose an innovative method dubbed Virtual Node Tuning (VNT). Our approach utilizes a pretrained graph transformer as the encoder and injects virtual nodes as soft prompts in the embedding space, which can be optimized with few-shot labels in novel classes to modulate node embeddings for each specific FSNC task. A unique feature of VNT is that, by incorporating a Graph-based Pseudo Prompt Evolution (GPPE) module, VNT-GPPE can handle scenarios with …

   @inproceedings{tan2023virtual,
     pubtype = {published},
     title = {Virtual Node Tuning for Few-shot Node Classification},
     author = {Tan, Zhen and Guo, Ruocheng and Ding, Kaize and Liu, Huan},
     booktitle = {Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
     year = {2023},
     url = {https://dl.acm.org/doi/abs/10.1145/3580305.3599541},
   }

3. Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance

   Song Wang, Zhen Tan, Ruocheng Guo, and 1 more author

   In EMNLP Findings, 2023

   Abs Bib

   Adopting a two-stage paradigm of pretraining followed by fine-tuning, Pretrained Language Models (PLMs) have achieved substantial advancements in the field of natural language processing. However, in real-world scenarios, data labels are often noisy due to the complex annotation process, making it essential to develop strategies for fine-tuning PLMs with such noisy labels. To this end, we introduce an innovative approach for fine-tuning PLMs using noisy labels, which incorporates the guidance of Large Language Models (LLMs) like ChatGPT. This guidance assists in accurately distinguishing between clean and noisy samples and provides supplementary information beyond the noisy labels, thereby boosting the learning process during fine-tuning PLMs. Extensive experiments on synthetic and real-world noisy datasets further demonstrate the superior advantages of our framework over the state-of-the-art baselines.

   @inproceedings{wang2023noise,
     pubtype = {published},
     title = {Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance},
     author = {Wang, Song and Tan, Zhen and Guo, Ruocheng and Li, Jundong},
     booktitle = {EMNLP Findings},
     year = {2023},
     url = {https://aclanthology.org/2023.findings-emnlp.834/},
   }

4. Debiasing Recommendation by Learning Identifiable Latent Confounders

   Qing Zhang, Xiaoying Zhang, Yang Liu, and 4 more authors

   In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2023

   Abs Bib

   Recommendation systems aim to predict users’ feedback on items not exposed to them yet. Confounding bias arises due to the presence of unmeasured variables (e.g., the socio-economic status of a user) that can affect both a user’s exposure and feedback. Existing methods either (1) make untenable assumptions about these unmeasured variables or (2) directly infer latent confounders from users’ exposure. However, they cannot guarantee the identification of counterfactual feedback, which can lead to biased predictions. In this work, we propose a novel method, i.e., identifiable deconfounder (iDCF), which leverages a set of proxy variables (e.g., observed user features) to resolve the aforementioned non-identification issue. The proposed iDCF is a general deconfounded recommendation framework that applies proximal causal inference to infer the unmeasured confounders and identify the counterfactual …

   @inproceedings{zhang2023debiasing,
     pubtype = {published},
     title = {Debiasing Recommendation by Learning Identifiable Latent Confounders},
     author = {Zhang, Qing and Zhang, Xiaoying and Liu, Yang and Wang, Hongning and Gao, Min and Zhang, Jiheng and Guo, Ruocheng},
     booktitle = {Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
     year = {2023},
     url = {https://dl.acm.org/doi/abs/10.1145/3580305.3599296},
   }

5. Learning for counterfactual fairness from observational data

   Jing Ma, Ruocheng Guo, Aidong Zhang, and 1 more author

   2023

   Abs Bib

   Fairness-aware machine learning has attracted a surge of attention in many domains, such as online advertising, personalized recommendation, and social media analysis in web applications. Fairness-aware machine learning aims to eliminate biases of learning models against certain subgroups described by certain protected (sensitive) attributes such as race, gender, and age. Among many existing fairness notions, counterfactual fairness is a popular notion defined from a causal perspective. It measures the fairness of a predictor by comparing the prediction of each individual in the original world and that in the counterfactual worlds in which the value of the sensitive attribute is modified. A prerequisite for existing methods to achieve counterfactual fairness is the prior human knowledge of the causal model for the data. However, in real-world scenarios, the underlying causal model is often unknown, and acquiring …

   @article{ma2023learning,
     pubtype = {published},
     title = {Learning for counterfactual fairness from observational data},
     author = {Ma, Jing and Guo, Ruocheng and Zhang, Aidong and Li, Jundong},
     pages = {1620-1630},
     year = {2023},
     url = {https://dl.acm.org/doi/abs/10.1145/3580305.3599408},
   }

6. Equal opportunity of coverage in fair regression

   Fangxin Wang, Lu Cheng, Ruocheng Guo, and 2 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2023

   Abs Bib

   We study fair machine learning (ML) under predictive uncertainty to enable reliable and trustworthy decision-making. The seminal work of’equalized coverage’proposed an uncertainty-aware fairness notion. However, it does not guarantee equal coverage rates across more fine-grained groups (eg, low-income females) conditioning on the true label and is biased in the assessment of uncertainty. To tackle these limitations, we propose a new uncertainty-aware fairness–Equal Opportunity of Coverage (EOC)–that aims to achieve two properties:(1) coverage rates for different groups with similar outcomes are close, and (2) the coverage rate for the entire population remains at a predetermined level. Further, the prediction intervals should be narrow to be informative. We propose Binned Fair Quantile Regression (BFQR), a distribution-free post-processing method to improve EOC with reasonable width for any trained ML models. It first calibrates a hold-out set to bound deviation from EOC, then leverages conformal prediction to maintain EOC on a test set, meanwhile optimizing prediction interval width. Experimental results demonstrate the effectiveness of our method in improving EOC.

   @inproceedings{wang2023equal,
     pubtype = {published},
     title = {Equal opportunity of coverage in fair regression},
     author = {Wang, Fangxin and Cheng, Lu and Guo, Ruocheng and Liu, Kay and Yu, Philip S},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2023},
     url = {https://proceedings.neurips.cc/paper_files/paper/2023/hash/1849b94ed817ae7043a6b6934ef410c1-Abstract-Conference.html},
   }

7. Causal disentanglement for implicit recommendations with network information

   Paras Sheth, Ruocheng Guo, Lu Cheng, and 2 more authors

   ACM Transactions on Knowledge Discovery from Data, 2023

   Abs Bib

   Online user engagement is highly influenced by various machine learning models, such as recommender systems. These systems recommend new items to the user based on the user’s historical interactions. Implicit recommender systems reflect a binary setting showing whether a user interacted (e.g., clicked on) with an item or not. However, the observed clicks may be due to various causes such as user’s interest, item’s popularity, and social influence factors. Traditional recommender systems consider these causes under a unified representation, which may lead to the emergence and amplification of various biases in recommendations. However, recent work indicates that by disentangling the unified representations, one can mitigate bias (e.g., popularity bias) in recommender systems and help improve recommendation performance. Yet, prior work in causal disentanglement in recommendations does not …

   @article{sheth2023causal,
     pubtype = {published},
     title = {Causal disentanglement for implicit recommendations with network information},
     author = {Sheth, Paras and Guo, Ruocheng and Cheng, Lu and Liu, Huan and Candan, Kasim Selçuk},
     journal = {ACM Transactions on Knowledge Discovery from Data},
     volume = {17},
     number = {7},
     pages = {1-18},
     year = {2023},
     publisher = {ACM},
     url = {https://dl.acm.org/doi/abs/10.1145/3582435},
   }

8. Modeling interference for individual treatment effect estimation from networked observational data

   Qiang Huang, Jing Ma, Jundong Li, and 3 more authors

   ACM Transactions on Knowledge Discovery from Data, 2023

   Abs Bib

   Estimating individual treatment effect (ITE) from observational data has attracted great interest in recent years, which plays a crucial role in decision-making across many high-impact domains such as economics, medicine, and e-commerce. Most existing studies of ITE estimation assume that different units at play are independent and do not influence each other. However, many social science experiments have shown that there often exist different levels of interactions between units in observational data, especially in a networked environment. As a result, the treatment assignment of one unit can affect the outcome of other units connected to it in the network, which is referred to as the interference or spillover effect. In this article, we study an important problem of ITE estimation from networked observational data by modeling the interference between different units and provide a principled framework to support such …

   @article{huang2023modeling,
     pubtype = {published},
     title = {Modeling interference for individual treatment effect estimation from networked observational data},
     author = {Huang, Qiang and Ma, Jing and Li, Jundong and Guo, Ruocheng and Sun, Huiyan and Chang, Yi},
     journal = {ACM Transactions on Knowledge Discovery from Data},
     volume = {18},
     number = {3},
     pages = {1-21},
     year = {2023},
     publisher = {ACM},
     url = {https://dl.acm.org/doi/abs/10.1145/3628449},
   }

9. Federated Knowledge Graph Completion via Latent Embedding Sharing and Tensor Factorization

   Maolin Wang, Dun Zeng, Zenglin Xu, and 2 more authors

   In Proceedings, 2023

   DOI Bib

   @inproceedings{wang2023federated,
     pubtype = {published},
     title = {Federated Knowledge Graph Completion via Latent Embedding Sharing and Tensor Factorization},
     author = {Wang, Maolin and Zeng, Dun and Xu, Zenglin and Guo, Ruocheng and Zhao, Xiangyu},
     booktitle = {Proceedings},
     pages = {1361--1366},
     year = {2023},
     doi = {10.1109/icdm58522.2023.00175}
   }

10. What Boosts Fake News Dissemination on Social Media? A Causal Inference View

    Yichuan Li, Kyumin Lee, Nima Kordzadeh, and 1 more author

    In Lecture notes in computer science, 2023

    DOI Bib

    @inproceedings{li2023what,
      pubtype = {published},
      title = {What Boosts Fake News Dissemination on Social Media? A Causal Inference View},
      author = {Li, Yichuan and Lee, Kyumin and Kordzadeh, Nima and Guo, Ruocheng},
      booktitle = {Lecture notes in computer science},
      pages = {234--246},
      year = {2023},
      doi = {10.1007/978-3-031-33383-5_19}
    }

11. Graph-Based Model-Agnostic Data Subsampling for Recommendation Systems

    Xiaohui Chen, Jiankai Sun, Taiqing Wang, and 3 more authors

    In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2023

    DOI arXiv Bib

    @inproceedings{chen2023graph,
      pubtype = {published},
      title = {Graph-Based Model-Agnostic Data Subsampling for Recommendation Systems},
      author = {Chen, Xiaohui and Sun, Jiankai and Wang, Taiqing and Guo, Ruocheng and Liu, Liping and Zhang, Aonan},
      booktitle = {ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)},
      pages = {3865--3876},
      year = {2023},
      doi = {10.1145/3580305.3599834},
    }

12. Causal Inference on Graphs

    Jing Ma, Ruocheng Guo, and Jundong Li

    Springer eBooks, 2023

    DOI Bib

    @article{ma2023causal,
      pubtype = {published},
      title = {Causal Inference on Graphs},
      author = {Ma, Jing and Guo, Ruocheng and Li, Jundong},
      journal = {Springer eBooks},
      pages = {53--78},
      year = {2023},
      doi = {10.1007/978-3-031-35051-1_4}
    }

2022

1. Learning Fair Node Representations with Graph Counterfactual Fairness

   Jing Ma, Ruocheng Guo, Mengting Wan, and 3 more authors

   In Proceedings of the 15th ACM International Conference on Web Search and Data Mining, 2022

   Abs Bib

   Fair machine learning aims to mitigate the biases of model predictions against certain subpopulations regarding sensitive attributes such as race and gender. Among the many existing fairness notions, counterfactual fairness measures the model fairness from a causal perspective by comparing the predictions of each individual from the original data and the counterfactuals. In counterfactuals, the sensitive attribute values of this individual had been modified. Recently, a few works extend counterfactual fairness to graph data, but most of them neglect the following facts that can lead to biases: 1) the sensitive attributes of each node’s neighbors may causally affect the prediction w.r.t. this node; 2) the sensitive attributes may causally affect other features and the graph structure. To tackle these issues, in this paper, we propose a novel fairness notion - graph counterfactual fairness, which considers the biases led by the …

   @inproceedings{ma2022learning,
     pubtype = {published},
     title = {Learning Fair Node Representations with Graph Counterfactual Fairness},
     author = {Ma, Jing and Guo, Ruocheng and Wan, Mengting and Yang, Longqi and Zhang, Aidong and Li, Jundong},
     booktitle = {Proceedings of the 15th ACM International Conference on Web Search and Data Mining},
     year = {2022},
     url = {https://dl.acm.org/doi/abs/10.1145/3488560.3498391},
   }

2. Evaluation methods and measures for causal learning algorithms

   Lu Cheng, Ruocheng Guo, Raha Moraffah, and 3 more authors

   IEEE, 2022

   Abs Bib

   The convenient access to copious multifaceted data has encouraged machine learning researchers to reconsider correlation-based learning and embrace the opportunity of causality-based learning, i.e., causal machine learning (causal learning). Recent years have, therefore, witnessed great effort in developing causal learning algorithms aiming to help artificial intelligence (AI) achieve human-level intelligence. Due to the lack of ground-truth data, one of the biggest challenges in current causal learning research is algorithm evaluations. This largely impedes the cross-pollination of AI and causal inference and hinders the two fields to benefit from the advances of the other. To bridge from conventional causal inference (i.e., based on statistical methods) to causal learning with Big Data (i.e., the intersection of causal inference and machine learning), in this survey, we review commonly used datasets, evaluation …

   @article{cheng2022evaluation,
     pubtype = {published},
     title = {Evaluation methods and measures for causal learning algorithms},
     author = {Cheng, Lu and Guo, Ruocheng and Moraffah, Raha and Sheth, Paras and Candan, K Selçuk and Liu, Huan},
     journal = {IEEE},
     volume = {3},
     number = {6},
     pages = {924-943},
     year = {2022},
     url = {https://ieeexplore.ieee.org/abstract/document/9709543/},
   }

3. CLEAR: Generative Counterfactual Explanations on Graphs

   Jing Ma, Ruocheng Guo, Saumitra Mishra, and 2 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2022

   Abs Bib

   Counterfactual explanations promote explainability in machine learning models by answering the question “how should the input instance be altered to obtain a desired predicted label?". The comparison of this instance before and after perturbation can enhance human interpretation. Most existing studies on counterfactual explanations are limited in tabular data or image data. In this paper, we study the problem of counterfactual explanation generation on graphs. A few studies have explored to generate counterfactual explanations on graphs, but many challenges of this problem are still not well-addressed: 1) optimizing in the discrete and disorganized space of graphs; 2) generalizing on unseen graphs; 3) maintaining the causality in the generated counterfactuals without prior knowledge of the causal model. To tackle these challenges, we propose a novel framework CLEAR which aims to generate counterfactual explanations on graphs for graph-level prediction models. Specifically, CLEAR leverages a graph variational autoencoder based mechanism to facilitate its optimization and generalization, and promotes causality by leveraging an auxiliary variable to better identify the causal model. Extensive experiments on both synthetic and real-world graphs validate the superiority of CLEAR over state-of-the-art counterfactual explanation methods on graphs in different aspects.

   @inproceedings{ma2022clear,
     pubtype = {published},
     title = {CLEAR: Generative Counterfactual Explanations on Graphs},
     author = {Ma, Jing and Guo, Ruocheng and Mishra, Saumitra and Zhang, Aidong and Li, Jundong},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2022},
     url = {https://proceedings.neurips.cc/paper_files/paper/2022/hash/a69d7f3a1340d55c720e572742439eaf-Abstract-Conference.html},
   }

4. MLP4Rec: A Pure MLP Architecture for Sequential Recommendations

   Muyang Li, Xiangyu Zhao, Chuan Lyu, and 3 more authors

   In International Joint Conference on Artificial Intelligence (IJCAI), 2022

   Abs Bib

   Self-attention models have achieved state-of-the-art performance in sequential recommender systems by capturing the sequential dependencies among user-item interactions. However, they rely on positional embeddings to retain the sequential information, which may break the semantics of item embeddings. In addition, most existing works assume that such sequential dependencies exist solely in the item embeddings, but neglect their existence among the item features. In this work, we propose a novel sequential recommender system (MLP4Rec) based on the recent advances of MLP-based architectures, which is naturally sensitive to the order of items in a sequence. To be specific, we develop a tri-directional fusion scheme to coherently capture sequential, cross-channel and cross-feature correlations. Extensive experiments demonstrate the effectiveness of MLP4Rec over various representative baselines upon two benchmark datasets. The simple architecture of MLP4Rec also leads to the linear computational complexity as well as much fewer model parameters than existing self-attention methods.

   @inproceedings{li2022mlp4rec,
     pubtype = {published},
     title = {MLP4Rec: A Pure MLP Architecture for Sequential Recommendations},
     author = {Li, Muyang and Zhao, Xiangyu and Lyu, Chuan and Zhao, Minghao and Wu, Runze and Guo, Ruocheng},
     booktitle = {International Joint Conference on Artificial Intelligence (IJCAI)},
     year = {2022},
     url = {https://arxiv.org/abs/2204.11510},
   }

5. Graph Few-shot Class-incremental Learning

   Zhen Tan, Kaize Ding, Ruocheng Guo, and 1 more author

   In Proceedings of the 15th ACM International Conference on Web Search and Data Mining, 2022

   Abs Bib

   The ability to incrementally learn new classes is vital to all real-world artificial intelligence systems. A large portion of high-impact applications like social media, recommendation systems, E-commerce platforms, etc. can be represented by graph models. In this paper, we investigate the challenging yet practical problem,Graph Few-shot Class-incremental (Graph FCL) problem, where the graph model is tasked to classify both newly encountered classes and previously learned classes. Towards that purpose, we put forward a Graph Pseudo Incremental Learning paradigm by sampling tasks recurrently from the base classes, so as to produce an arbitrary number of training episodes for our model to practice the incremental learning skill. Furthermore, we design a Hierarchical-Attention-based Graph Meta-learning framework, HAG-Meta from an optimization perspective. We present a task-sensitive regularizer calculated …

   @inproceedings{tan2022graph,
     pubtype = {published},
     title = {Graph Few-shot Class-incremental Learning},
     author = {Tan, Zhen and Ding, Kaize and Guo, Ruocheng and Liu, Huan},
     booktitle = {Proceedings of the 15th ACM International Conference on Web Search and Data Mining},
     year = {2022},
     url = {https://dl.acm.org/doi/abs/10.1145/3488560.3498455},
   }

6. Accurate identification of bacteriophages from metagenomic data using Transformer

   Jiayu Shang, Xubo Tang, Ruocheng Guo, and 1 more author

   Briefings in Bioinformatics, Volume 23, Issue 4, July 2022, 2022

   Abs Bib

   Bacteriophages are viruses infecting bacteria. Being key players in microbial communities, they can regulate the composition/function of microbiome by infecting their bacterial hosts and mediating gene transfer. Recently, metagenomic sequencing, which can sequence all genetic materials from various microbiome, has become a popular means for new phage discovery. However, accurate and comprehensive detection of phages from the metagenomic data remains difficult. High diversity/abundance, and limited reference genomes pose major challenges for recruiting phage fragments from metagenomic data. Existing alignment-based or learning-based models have either low recall or precision on metagenomic data.In this work, we adopt the state-of-the-art language model, Transformer, to conduct contextual embedding for phage contigs. By constructing …

   @article{shang2022accurate,
     pubtype = {published},
     title = {Accurate identification of bacteriophages from metagenomic data using Transformer},
     author = {Shang, Jiayu and Tang, Xubo and Guo, Ruocheng and Sun, Yanni},
     journal = {Briefings in Bioinformatics, Volume 23, Issue 4, July 2022},
     year = {2022},
     url = {https://academic.oup.com/bib/article-abstract/23/4/bbac258/6620872},
   }

7. Nothing Stands Alone: Relational Fake News Detection with Hypergraph Neural Networks

   Ujun Jeong, Kaize Ding, Lu Cheng, and 3 more authors

   In The IEEE International Conference on Big Data (BigData), 2022

   Abs Bib

   Nowadays, fake news easily propagates through online social networks and becomes a grand threat to individuals and society. Assessing the authenticity of news is challenging due to its elaborately fabricated contents, making it difficult to obtain large-scale annotations for fake news data. Due to such data scarcity issues, detecting fake news tends to fail and overfit in the supervised setting. Recently, graph neural networks (GNNs) have been adopted to leverage the richer relational information among both labeled and unlabeled instances. Despite their promising results, they are inherently focused on pairwise relations between news, which can limit the expressive power for capturing fake news that spreads in a group-level. For example, detecting fake news can be more effective when we better understand relations between news pieces shared among susceptible users. To address those issues, we propose to …

   @inproceedings{jeong2022nothing,
     pubtype = {published},
     title = {Nothing Stands Alone: Relational Fake News Detection with Hypergraph Neural Networks},
     author = {Jeong, Ujun and Ding, Kaize and Cheng, Lu and Guo, Ruocheng and Shu, Kai and Liu, Huan},
     booktitle = {The IEEE International Conference on Big Data (BigData)},
     year = {2022},
     url = {https://ieeexplore.ieee.org/abstract/document/10020234/},
   }

8. Mitigating popularity bias in recommendation with unbalanced interactions: A gradient perspective

   Weijieying Ren, Lei Wang, Kunpeng Liu, and 3 more authors

   In IEEE International Conference on Data Mining (ICDM), 2022

   Abs Bib

   Recommender systems learn from historical user-item interactions to identify preferred items for target users. These observed interactions are usually unbalanced following a long-tailed distribution. Such long-tailed data lead to popularity bias to recommend popular but not personalized items to users. We present a gradient perspective to understand two negative impacts of popularity bias in recommendation model optimization: (i) the gradient direction of popular item embeddings is closer to that of positive interactions, and (ii) the magnitude of positive gradient for popular items are much greater than that of unpopular items. To address these issues, we propose a simple yet efficient framework to mitigate popularity bias from a gradient perspective. Specifically, we first normalize each user embedding and record accumulated gradients of users and items via popularity bias measures in model training. To address …

   @inproceedings{ren2022mitigating,
     pubtype = {published},
     title = {Mitigating popularity bias in recommendation with unbalanced interactions: A gradient perspective},
     author = {Ren, Weijieying and Wang, Lei and Liu, Kunpeng and Guo, Ruocheng and Peng, Lim Ee and Fu, Yanjie},
     booktitle = {IEEE International Conference on Data Mining (ICDM)},
     year = {2022},
     url = {https://ieeexplore.ieee.org/abstract/document/10027732/},
   }

9. Causal mediation analysis with hidden confounders

   Lu Cheng, Ruocheng Guo, and Huan Liu

   2022

   Abs Bib

   An important problem in causal inference is to break down the total effect of a treatment on an outcome into different causal pathways and to quantify the causal effect in each pathway. For instance, in causal fairness, the total effect of being a male employee (i.e., treatment) constitutes its direct effect on annual income (i.e., outcome) and the indirect effect via the employee’s occupation (i.e., mediator). Causal mediation analysis (CMA) is a formal statistical framework commonly used to reveal such underlying causal mechanisms. One major challenge of CMA in observational studies is handling confounders, variables that cause spurious causal relationships among treatment, mediator, and outcome. Conventional methods assume sequential ignorability that implies all confounders can be measured, which is often unverifiable in practice. This work aims to circumvent the stringent sequential ignorability assumptions …

   @article{cheng2022causal,
     pubtype = {published},
     title = {Causal mediation analysis with hidden confounders},
     author = {Cheng, Lu and Guo, Ruocheng and Liu, Huan},
     pages = {113-122},
     year = {2022},
     url = {https://dl.acm.org/doi/abs/10.1145/3488560.3498407},
   }

10. Supervised Graph Contrastive Learning for Few-shot Node Classification

    Zhen Tan, Kaize Ding, Ruocheng Guo, and 1 more author

    ECML-PKDD 2022, 2022

    Abs Bib

    Graphs present in many real-world applications, such as financial fraud detection, commercial recommendation, and social network analysis. But given the high cost of graph annotation or labeling, we face a severe graph label-scarcity problem, i.e., a graph might have a few labeled nodes. One example of such a problem is the so-called few-shot node classification. A predominant approach to this problem resorts to episodic meta-learning. In this work, we challenge the status quo by asking a fundamental question whether meta-learning is a must for few-shot node classification tasks. We propose a new and simple framework under the standard few-shot node classification setting as an alternative to meta-learning to learn an effective graph encoder. The framework consists of supervised graph contrastive learning with novel mechanisms for data augmentation, subgraph encoding, and multi-scale contrast on …

    @article{tan2022supervised,
      pubtype = {published},
      title = {Supervised Graph Contrastive Learning for Few-shot Node Classification},
      author = {Tan, Zhen and Ding, Kaize and Guo, Ruocheng and Liu, Huan},
      journal = {ECML-PKDD 2022},
      year = {2022},
      url = {https://link.springer.com/chapter/10.1007/978-3-031-26390-3_24},
    }

11. Causal Disentanglement with Network Information for Debiased Recommendations

    Paras Sheth, Ruocheng Guo, Kaize Ding, and 3 more authors

    In Lecture notes in computer science, 2022

    DOI Bib

    @inproceedings{sheth2022causal,
      pubtype = {published},
      title = {Causal Disentanglement with Network Information for Debiased Recommendations},
      author = {Sheth, Paras and Guo, Ruocheng and Ding, Kaize and Cheng, Lu and Candan, K. Selçuk and Liu, Huan},
      booktitle = {Lecture notes in computer science},
      pages = {265--273},
      year = {2022},
      doi = {10.1007/978-3-031-17849-8_21}
    }

12. Effects of Multi-Aspect Online Reviews with Unobserved Confounders: Estimation and Implication

    Lu Cheng, Ruocheng Guo, Kasim Candan, and 1 more author

    In Proceedings of the International AAAI Conference on Web and Social Media, 2022

    DOI Bib

    @inproceedings{cheng2022effects,
      pubtype = {published},
      title = {Effects of Multi-Aspect Online Reviews with Unobserved Confounders: Estimation and Implication},
      author = {Cheng, Lu and Guo, Ruocheng and Candan, Kasim and Liu, Huan},
      booktitle = {Proceedings of the International AAAI Conference on Web and Social Media},
      volume = {16},
      pages = {67--78},
      year = {2022},
      doi = {10.1609/icwsm.v16i1.19273}
    }

13. Distributional Shift Adaptation using Domain-Specific Features

    Anique Tahir, Cheng Zhu Lu, Ruocheng Guo, and 1 more author

    In IEEE International Conference on Big Data (BigData), 2022

    DOI Bib

    @inproceedings{tahir2022distributional,
      pubtype = {published},
      title = {Distributional Shift Adaptation using Domain-Specific Features},
      author = {Tahir, Anique and Lu, Cheng Zhu and Guo, Ruocheng and Liu, Huan},
      booktitle = {IEEE International Conference on Big Data (BigData)},
      pages = {5593--5597},
      year = {2022},
      doi = {10.1109/bigdata55660.2022.10020444}
    }

14. Estimating Causal Effects of Multi-Aspect Online Reviews with Multi-Modal Proxies

    Lu Cheng, Ruocheng Guo, and Huan Liu

    In ACM International Conference on Web Search and Data Mining (WSDM), 2022

    DOI Bib

    @inproceedings{cheng2022estimating,
      pubtype = {published},
      title = {Estimating Causal Effects of Multi-Aspect Online Reviews with Multi-Modal Proxies},
      author = {Cheng, Lu and Guo, Ruocheng and Liu, Huan},
      booktitle = {ACM International Conference on Web Search and Data Mining (WSDM)},
      pages = {103--112},
      year = {2022},
      doi = {10.1145/3488560.3498372}
    }

2021

1. Causal understanding of fake news dissemination on social media

   Lu Cheng, Ruocheng Guo, Kai Shu, and 1 more author

   2021

   Abs Bib

   Recent years have witnessed remarkable progress towards computational fake news detection. To mitigate its negative impact, we argue that it is critical to understand what user attributes potentially cause users to share fake news. The key to this causal-inference problem is to identify confounders – variables that cause spurious associations between treatments (e.g., user attributes) and outcome (e.g., user susceptibility). In fake news dissemination, confounders can be characterized by fake news sharing behavior that inherently relates to user attributes and online activities. Learning such user behavior is typically subject to selection bias in users who are susceptible to share news on social media. Drawing on causal inference theories, we first propose a principled approach to alleviating selection bias in fake news dissemination. We then consider the learned unbiased fake news sharing behavior as the …

   @article{cheng2021causal,
     pubtype = {published},
     title = {Causal understanding of fake news dissemination on social media},
     author = {Cheng, Lu and Guo, Ruocheng and Shu, Kai and Liu, Huan},
     pages = {148-157},
     year = {2021},
     url = {https://dl.acm.org/doi/abs/10.1145/3447548.3467321},
   }

2. Adversarial machine learning: An interpretation perspective

   Ninghao Liu, Mengnan Du, Ruocheng Guo, and 2 more authors

   ACM SIGKDD Explorations Newsletter, 2021

   Bib

   @article{liu2021adversarial,
     pubtype = {published},
     title = {Adversarial machine learning: An interpretation perspective},
     author = {Liu, Ninghao and Du, Mengnan and Guo, Ruocheng and Liu, Huan and Hu, Xia},
     journal = {ACM SIGKDD Explorations Newsletter},
     year = {2021},
   }

3. Deconfounding with networked observational data in a dynamic environment

   Jing Ma, Ruocheng Guo, Chen Chen, and 2 more authors

   2021

   Abs Bib

   One fundamental problem in causal inference is to learn the individual treatment effects (ITE) – assessing the causal effects of a certain treatment (e.g., prescription of medicine) on an important outcome (e.g., cure of a disease) for each data instance, but the effectiveness of most existing methods is often limited due to the existence of hidden confounders. Recent studies have shown that the auxiliary relational information among data can be utilized to mitigate the confounding bias. However, these works assume that the observational data and the relations among them are static, while in reality, both of them will continuously evolve over time and we refer such data as time-evolving networked observational data.In this paper, we make an initial investigation of ITE estimation on such data. The problem remains difficult due to the following challenges: (1) modeling the evolution patterns of time-evolving networked …

   @article{ma2021deconfounding,
     pubtype = {published},
     title = {Deconfounding with networked observational data in a dynamic environment},
     author = {Ma, Jing and Guo, Ruocheng and Chen, Chen and Zhang, Aidong and Li, Jundong},
     pages = {166-174},
     year = {2021},
     url = {https://dl.acm.org/doi/abs/10.1145/3437963.3441818},
   }

4. Modeling temporal patterns of cyberbullying detection with hierarchical attention networks

   Lu Cheng, Ruocheng Guo, Yasin N Silva, and 2 more authors

   ACM/IMS Transactions on Data Science, 2021

   Abs Bib

   Cyberbullying is rapidly becoming one of the most serious online risks for adolescents. This has motivated work on machine learning methods to automate the process of cyberbullying detection, which have so far mostly viewed cyberbullying as one-off incidents that occur at a single point in time. Comparatively less is known about how cyberbullying behavior occurs and evolves over time. This oversight highlights a crucial open challenge for cyberbullying-related research, given that cyberbullying is typically defined as intentional acts of aggression via electronic communication that occur repeatedly and persistently. In this article, we center our discussion on the challenge of modeling temporal patterns of cyberbullying behavior. Specifically, we investigate how temporal information within a social media session, which has an inherently hierarchical structure (e.g., words form a comment and comments form a …

   @article{cheng2021modeling,
     pubtype = {published},
     title = {Modeling temporal patterns of cyberbullying detection with hierarchical attention networks},
     author = {Cheng, Lu and Guo, Ruocheng and Silva, Yasin N and Hall, Deborah and Liu, Huan},
     journal = {ACM/IMS Transactions on Data Science},
     volume = {2},
     number = {2},
     pages = {1-23},
     year = {2021},
     publisher = {ACM},
     url = {https://dl.acm.org/doi/abs/10.1145/3441141},
   }

5. Ignite: A minimax game toward learning individual treatment effects from networked observational data

   Ruocheng Guo, Jundong Li, Yichuan Li, and 3 more authors

   2021

   Abs Bib

   Networked observational data presents new opportunities for learning individual causal effects, which plays an indispensable role in decision making. Such data poses the challenge of confounding bias. Previous work presents two desiderata to handle confounding bias. On the treatment group level, we aim to balance the distributions of confounder representations. On the individual level, it is desirable to capture patterns of hidden confounders that predict treatment assignments. Existing methods show the potential of utilizing network information to handle confounding bias, but they only try to satisfy one of the two desiderata. This is because the two desiderata seem to contradict each other. When the two distributions of confounder representations are highly overlapped, then we confront the undiscriminating problem between the treated and the controlled. In this work, we formulate the two desiderata as a minimax game. We propose IGNITE that learns representations of confounders from networked observational data, which is trained by a minimax game to achieve the two desiderata. Experiments verify the efficacy of IGNITE on two datasets under various settings.

   @article{guo2021ignite,
     pubtype = {published},
     title = {Ignite: A minimax game toward learning individual treatment effects from networked observational data},
     author = {Guo, Ruocheng and Li, Jundong and Li, Yichuan and Candan, K Selçuk and Raglin, Adrienne and Liu, Huan},
     pages = {4534-4540},
     year = {2021},
     url = {https://www.ijcai.org/proceedings/2020/0625.pdf},
   }

6. Long-term effect estimation with surrogate representation

   Lu Cheng, Ruocheng Guo, and Huan Liu

   2021

   Abs Bib

   There are many scenarios where short- and long-term causal effects of an intervention are different. For example, low-quality ads may increase short-term ad clicks but decrease the long-term revenue via reduced clicks. This work, therefore, studies the the problem of long-term effect where the outcome of primary interest, orprimary outcome, takes months or even years to accumulate. The observational study of long-term effect presents unique challenges. First, the confounding bias causes large estimation error and variance, which can further accumulate towards the prediction of primary outcomes. Second, short-term outcomes are often directly used as the proxy of the primary outcome, i.e., thesurrogate. Nevertheless, this method entails the strong surrogacy assumption that is often impractical. To tackle these challenges, we propose to build connections between long-term causal inference and sequential models …

   @article{cheng2021long,
     pubtype = {published},
     title = {Long-term effect estimation with surrogate representation},
     author = {Cheng, Lu and Guo, Ruocheng and Liu, Huan},
     pages = {274-282},
     year = {2021},
     url = {https://dl.acm.org/doi/abs/10.1145/3437963.3441719},
   }

7. Multi-Cause Effect Estimation with Disentangled Confounder Representation

   Jing Ma, Ruocheng Guo, Aidong Zhang, and 1 more author

   In , 2021

   DOI Bib

   @inproceedings{ma2021multi,
     pubtype = {published},
     title = {Multi-Cause Effect Estimation with Disentangled Confounder Representation},
     author = {Ma, Jing and Guo, Ruocheng and Zhang, Aidong and Li, Jundong},
     pages = {2790--2796},
     year = {2021},
     doi = {10.24963/ijcai.2021/384}
   }

8. CauseBox

   Paras Sheth, Ujun Jeong, Ruocheng Guo, and 2 more authors

   In Proceedings of the ... ACM International Conference on Information & Knowledge Management, 2021

   DOI Bib

   @inproceedings{sheth2021causebox,
     pubtype = {published},
     title = {CauseBox},
     author = {Sheth, Paras and Jeong, Ujun and Guo, Ruocheng and Liu, Huan and Candan, K. Selçuk},
     booktitle = {Proceedings of the ... ACM International Conference on Information \& Knowledge Management},
     pages = {4789--4793},
     year = {2021},
     doi = {10.1145/3459637.3481974}
   }

2020

1. A survey of learning causality with data: Problems and methods

   Ruocheng Guo, Lu Cheng, Jundong Li, and 2 more authors

   ACM, 2020

   Abs Bib

   This work considers the question of how convenient access to copious data impacts our ability to learn causal effects and relations. In what ways is learning causality in the era of big data different from—or the same as—the traditional one? To answer this question, this survey provides a comprehensive and structured review of both traditional and frontier methods in learning causality and relations along with the connections between causality and machine learning. This work points out on a case-by-case basis how big data facilitates, complicates, or motivates each approach.

   @article{guo2020survey,
     pubtype = {published},
     title = {A survey of learning causality with data: Problems and methods},
     author = {Guo, Ruocheng and Cheng, Lu and Li, Jundong and Hahn, P Richard and Liu, Huan},
     journal = {ACM},
     volume = {53},
     number = {4},
     pages = {1-37},
     year = {2020},
     url = {https://dl.acm.org/doi/abs/10.1145/3397269},
   }

2. Causal Interpretability for Machine Learning–Problems, Methods and Evaluation

   Raha Moraffah, Mansooreh Karami, Ruocheng Guo, and 2 more authors

   ACM SIGKDD Explorations Newsletter, 2020

   Abs Bib

   Machine learning models have had discernible achievements in a myriad of applications. However, most of these models are black-boxes, and it is obscure how the decisions are made by them. This makes the models unreliable and untrustworthy. To provide insights into the decision making processes of these models, a variety of traditional interpretable models have been proposed. Moreover, to generate more humanfriendly explanations, recent work on interpretability tries to answer questions related to causality such as "Why does this model makes such decisions?" or "Was it a specific feature that caused the decision made by the model?". In this work, models that aim to answer causal questions are referred to as causal interpretable models. The existing surveys have covered concepts and methodologies of traditional interpretability. In this work, we present a comprehensive survey on causal interpretable …

   @article{moraffah2020causal,
     pubtype = {published},
     title = {Causal Interpretability for Machine Learning--Problems, Methods and Evaluation},
     author = {Moraffah, Raha and Karami, Mansooreh and Guo, Ruocheng and Raglin, Adrienne and Liu, Huan},
     journal = {ACM SIGKDD Explorations Newsletter},
     year = {2020},
     url = {https://dl.acm.org/doi/abs/10.1145/3400051.3400058},
   }

3. Learning Individual Causal Effects from Networked Observational Data

   Ruocheng Guo, Jundong Li, and Huan Liu

   In Proceedings of the 13th ACM International Conference on Web Search and Data Mining, 2020

   Abs Bib

   The convenient access to observational data enables us to learn causal effects without randomized experiments. This research direction draws increasing attention in research areas such as economics, healthcare, and education. For example, we can study how a medicine (the treatment) causally affects the health condition (the outcome) of a patient using existing electronic health records. To validate causal effects learned from observational data, we have to control confounding bias – the influence of variables which causally influence both the treatment and the outcome. Existing work along this line overwhelmingly relies on the unconfoundedness assumption that there do not exist unobserved confounders. However, this assumption is untestable and can even be untenable. In fact, an important fact ignored by the majority of previous work is that observational data can come with network information that can be …

   @inproceedings{guo2020learning,
     pubtype = {published},
     title = {Learning Individual Causal Effects from Networked Observational Data},
     author = {Guo, Ruocheng and Li, Jundong and Liu, Huan},
     booktitle = {Proceedings of the 13th ACM International Conference on Web Search and Data Mining},
     year = {2020},
     url = {https://dl.acm.org/doi/abs/10.1145/3336191.3371816},
   }

4. Privacy-aware recommendation with private-attribute protection using adversarial learning

   Ghazaleh Beigi, Ahmadreza Mosallanezhad, Ruocheng Guo, and 3 more authors

   2020

   Abs Bib

   Recommendation is one of the critical applications that helps users find information relevant to their interests. However, a malicious attacker can infer users’ private information via recommendations. Prior work obfuscates user-item data before sharing it with recommendation system. This approach does not explicitly address the quality of recommendation while performing data obfuscation. Moreover, it cannot protect users against private-attribute inference attacks based on recommendations. This work is the first attempt to build a Recommendation with Attribute Protection (RAP) model which simultaneously recommends relevant items and counters private-attribute inference attacks. The key idea of our approach is to formulate this problem as an adversarial learning problem with two main components: the private attribute inference attacker, and the Bayesian personalized recommender. The attacker seeks to infer …

   @article{beigi2020privacy,
     pubtype = {published},
     title = {Privacy-aware recommendation with private-attribute protection using adversarial learning},
     author = {Beigi, Ghazaleh and Mosallanezhad, Ahmadreza and Guo, Ruocheng and Alvari, Hamidreza and Nou, Alexander and Liu, Huan},
     pages = {34-42},
     year = {2020},
     url = {https://dl.acm.org/doi/abs/10.1145/3336191.3371832},
   }

5. Debiasing grid-based product search in e-commerce

   Ruocheng Guo, Xiaoting Zhao, Adam Henderson, and 2 more authors

   2020

   Abs Bib

   The widespread usage of e-commerce websites in daily life and the resulting wealth of implicit feedback data form the foundation for systems that train and test e-commerce search ranking algorithms. While convenient to collect, implicit feedback data inherently suffers from various types of bias since user feedback is limited to products they are exposed to by existing search ranking algorithms and impacted by how the products are displayed. In the literature, a vast majority of existing methods have been proposed towards unbiased learning to rank for list-based web search scenarios. However, such methods cannot be directly adopted by e-commerce websites mainly for two reasons. First, in e-commerce websites, search engine results pages (SERPs) are displayed in 2-dimensional grids. The existing methods have not considered the difference in user behavior between list-based web search and grid-based …

   @article{guo2020debiasing,
     pubtype = {published},
     title = {Debiasing grid-based product search in e-commerce},
     author = {Guo, Ruocheng and Zhao, Xiaoting and Henderson, Adam and Hong, Liangjie and Liu, Huan},
     pages = {2852-2860},
     year = {2020},
     url = {https://dl.acm.org/doi/abs/10.1145/3394486.3403336},
   }

6. Counterfactual evaluation of treatment assignment functions with networked observational data

   Ruocheng Guo, Jundong Li, and Huan Liu

   Society for Industrial and Applied Mathematics, 2020

   Abs Bib

   Counterfactual evaluation of novel treatment assignment functions (e.g., advertising algorithms and recommender systems) is one of the most crucial causal inference problems for practitioners. Traditionally, randomized controlled trials (A/B tests) are performed to evaluate treatment assignment functions. However, such trials can be time-consuming, expensive, and even unethical in some cases. Therefore, offline counterfactual evaluation of treatment assignment functions becomes a pressing issue because a massive amount of observational data is available in today’s big data era. Counterfactual evaluation requires handling the hidden confounders – the unmeasured features which causally influence both the treatment assignment and the outcome. To deal with the hidden confounders, most of the existing methods rely on the assumption of no hidden confounders. However, this assumption can be untenable in the context of massive observational data. When such data comes with network information, the later can be potentially useful to correct hidden confounding bias. As such, we first formulate a novel problem, counterfactual evaluation of treatment assignment functions with networked observational data. Then, we investigate the following research questions: How can we utilize network information in counterfactual evaluation? Can network information improve the estimates in counterfactual evaluation? Toward answering these questions, first, we propose a novel framework, \emphCounterfactual Network Evaluator (CONE), which (1) learns partial representations of latent confounders under the supervision of observed treatments and …

   @article{guo2020counterfactual,
     pubtype = {published},
     title = {Counterfactual evaluation of treatment assignment functions with networked observational data},
     author = {Guo, Ruocheng and Li, Jundong and Liu, Huan},
     journal = {Society for Industrial and Applied Mathematics},
     pages = {271-279},
     year = {2020},
     url = {https://arxiv.org/abs/1912.10536},
   }

7. Representation Learning for Imbalanced Cross-Domain Classification

   Lu Cheng, Ruocheng Guo, K. Selçuk Candan, and 1 more author

   In Society for Industrial and Applied Mathematics eBooks, 2020

   DOI Bib

   @inproceedings{cheng2020representation,
     pubtype = {published},
     title = {Representation Learning for Imbalanced Cross-Domain Classification},
     author = {Cheng, Lu and Guo, Ruocheng and Candan, K. Selçuk and Liu, Huan},
     booktitle = {Society for Industrial and Applied Mathematics eBooks},
     pages = {478--486},
     year = {2020},
     doi = {10.1137/1.9781611976236.54}
   }

8. A Practical Data Repository for Causal Learning with Big Data

   Lu Cheng, Ruocheng Guo, Raha Moraffah, and 3 more authors

   In Lecture notes in computer science, 2020

   DOI Bib

   @inproceedings{cheng2020practical,
     pubtype = {published},
     title = {A Practical Data Repository for Causal Learning with Big Data},
     author = {Cheng, Lu and Guo, Ruocheng and Moraffah, Raha and Candan, K. Selçuk and Raglin, Adrienne J. and Liu, Huan},
     booktitle = {Lecture notes in computer science},
     pages = {234--248},
     year = {2020},
     doi = {10.1007/978-3-030-49556-5_23}
   }

9. Causal Learning in Question Quality Improvement

   Yichuan Li, Ruocheng Guo, Weiying Wang, and 1 more author

   In Lecture notes in computer science, 2020

   DOI Bib

   @inproceedings{li2020causal,
     pubtype = {published},
     title = {Causal Learning in Question Quality Improvement},
     author = {Li, Yichuan and Guo, Ruocheng and Wang, Weiying and Liu, Huan},
     booktitle = {Lecture notes in computer science},
     pages = {204--214},
     year = {2020},
     doi = {10.1007/978-3-030-49556-5_20}
   }

10. Measuring Time-Constrained Influence to Predict Adoption in Online Social Networks

    Ericsson Marin, Ruocheng Guo, and Paulo Shakarian

    ACM Transactions on Social Computing, 2020

    DOI Bib

    @article{marin2020measuring,
      pubtype = {published},
      title = {Measuring Time-Constrained Influence to Predict Adoption in Online Social Networks},
      author = {Marin, Ericsson and Guo, Ruocheng and Shakarian, Paulo},
      journal = {ACM Transactions on Social Computing},
      volume = {3},
      number = {3},
      pages = {1--26},
      year = {2020},
      doi = {10.1145/3372785}
    }

2019

1. Hierarchical Attention Networks for Cyberbullying Detection on the Instagram Social Network

   Lu Cheng, Ruocheng Guo, Yasin Silva, and 2 more authors

   In Proceedings of the 2019 SIAM International Conference on Data Mining, 2019

   Abs Bib

   Cyberbullying has become one of the most pressing online risks for young people and has raised serious concerns in society. The emerging literature identifies cyberbullying as repetitive acts that occur over time rather than one-off incidents. Yet, there has been relatively little work to model the hierarchical structure of social media sessions and the temporal dynamics of cyberbullying in online social network sessions. We propose a hierarchical attention network for cyberbullying detection that takes these aspects of cyberbullying into account. The primary distinctive characteristics of our approach include: (i) a hierarchical structure that mirrors the structure of a social media session; (ii) levels of attention mechanisms applied at the word and comment level, thereby enabling the model to pay different amounts of attention to words and comments, depending on the context; and (iii) a cyberbullying detection task that also predicts the interval of time between two adjacent comments. These characteristics allow the model to exploit the commonalities and differences across these two tasks to improve the performance of cyberbullying detection. Experiments on a real-world dataset from Instagram, the social media platform on which the highest percentage of users have reported experiencing cyberbullying, reveal that the proposed architecture outperforms the state-of-the-art method.

   @inproceedings{cheng2019hierarchical,
     pubtype = {published},
     title = {Hierarchical Attention Networks for Cyberbullying Detection on the Instagram Social Network},
     author = {Cheng, Lu and Guo, Ruocheng and Silva, Yasin and Hall, Deborah and Liu, Huan},
     booktitle = {Proceedings of the 2019 SIAM International Conference on Data Mining},
     year = {2019},
     url = {https://par.nsf.gov/biblio/10110255},
   }

2. Multi-level network embedding with boosted low-rank matrix approximation

   Jundong Li, Liang Wu, Ruocheng Guo, and 2 more authors

   2019

   Abs Bib

   As opposed to manual feature engineering which is tedious and difficult to scale, network embedding has attracted a surge of research interests as it automates the feature learning on graphs. The learned low-dimensional node vectors ease the knowledge discovery on graphs by enabling various off-the-shelf machine learning tools to be directly applied. Recent research has shown that the past decade of network embedding approaches either explicitly factorize a carefully designed matrix or are closely related to implicit matrix factorization, with the fundamental assumption that the factorized node connectivity matrix is low-rank. Nonetheless, the global low-rank assumption does not necessarily hold especially when the factorized matrix encodes complex node interactions, and the resultant single low-rank embedding matrix is insufficient to capture all the observed connectivity patterns. In this regard, we propose a …

   @article{li2019multi,
     pubtype = {published},
     title = {Multi-level network embedding with boosted low-rank matrix approximation},
     author = {Li, Jundong and Wu, Liang and Guo, Ruocheng and Liu, Chenghao and Liu, Huan},
     pages = {49-56},
     year = {2019},
     url = {https://dl.acm.org/doi/abs/10.1145/3341161.3342864},
   }

3. Privacy preserving text representation learning

   Ghazaleh Beigi, Kai Shu, Ruocheng Guo, and 2 more authors

   2019

   Abs Bib

   Online users generate tremendous amounts of textual information by participating in different online activities. This data provides opportunities for researchers and business partners to understand individuals. However, this user-generated textual data not only can reveal the identity of the user but also may contain individual’s private attribute information. Publishing the textual data thus compromises the privacy of users. It is challenging to design effective anonymization techniques for textual information which minimize the chances of re-identification and does not contain private information while retaining the textual semantic meaning. In this paper, we study this problem and propose a novel double privacy preserving text representation learning framework, DPText. We show the effectiveness of DPText in preserving privacy and utility.

   @article{beigi2019privacy,
     pubtype = {published},
     title = {Privacy preserving text representation learning},
     author = {Beigi, Ghazaleh and Shu, Kai and Guo, Ruocheng and Wang, Suhang and Liu, Huan},
     pages = {275-276},
     year = {2019},
     url = {https://dl.acm.org/doi/abs/10.1145/3342220.3344925},
   }

4. Adaptive unsupervised feature selection on attributed networks

   Jundong Li, Ruocheng Guo, Chenghao Liu, and 1 more author

   2019

   Abs Bib

   Attributed networks are pervasive in numerous of high-impact domains. As opposed to conventional plain networks where only pairwise node dependencies are observed, both the network topology and node attribute information are readily available on attributed networks. More often than not, the nodal attributes are depicted in a high-dimensional feature space and are therefore notoriously difficult to tackle due to the curse of dimensionality. Additionally, features that are irrelevant to the network structure could hinder the discovery of actionable patterns from attributed networks. Hence, it is important to leverage feature selection to find a high-quality feature subset that is tightly correlated to the network structure. Few of the existing efforts either model the network structure at a macro-level by community analysis or directly make use of the binary relations. Consequently, they fail to exploit the finer-grained tie strength …

   @article{li2019adaptive,
     pubtype = {published},
     title = {Adaptive unsupervised feature selection on attributed networks},
     author = {Li, Jundong and Guo, Ruocheng and Liu, Chenghao and Liu, Huan},
     pages = {92-100},
     year = {2019},
     url = {https://dl.acm.org/doi/abs/10.1145/3292500.3330856},
   }

5. Robust cyberbullying detection with causal interpretation

   Lu Cheng, Ruocheng Guo, and Huan Liu

   2019

   Abs Bib

   Cyberbullying poses serious threats to preteens and teenagers, therefore, understanding the incentives behind cyberbullying is critical to prevent its happening and mitigate the impact. Most existing work towards cyberbullying detection has focused on the accuracy, and overlooked causes of the outcome. Discovering the causes of cyberbullying from observational data is challenging due to the existence of confounders, variables that can lead to spurious causal relationships between covariates and the outcome. This work studies the problem of robust cyberbullying detection with causal interpretation and proposes a principled framework to identify and block the influence of the plausible confounders, i.e., p-confounders. The de-confounded model is causally interpretable and is more robust to the changes in data distribution. We test our approach using the state-of-the-art evaluation method, causal transportability …

   @article{cheng2019robust,
     pubtype = {published},
     title = {Robust cyberbullying detection with causal interpretation},
     author = {Cheng, Lu and Guo, Ruocheng and Liu, Huan},
     pages = {169-175},
     year = {2019},
     url = {https://dl.acm.org/doi/abs/10.1145/3308560.3316503},
   }

6. Using network motifs to characterize temporal network evolution leading to diffusion inhibition: S. Sarkar et al.

   Soumajyoti Sarkar, Ruocheng Guo, and Paulo Shakarian

   Social Network Analysis and Mining, 2019

   Abs Bib

   Network motifs are patterns of over-represented node interactions in a network which have been previously used as building blocks to understand various aspects of the social networks. In this paper, we use motif patterns to characterize the information diffusion process in social networks. We study the lifecycle of information cascades to understand what leads to saturation of growth in terms of cascade reshares, thereby resulting in expiration, an event we call “diffusion inhibition”. In an attempt to understand what causes inhibition, we use motifs to dissect the network obtained from information cascades coupled with traces of historical diffusion or social network links. Our main results follow from experiments on a dataset of cascades from the Weibo platform and the Flixster movie ratings. We observe the temporal counts of 5-node undirected motifs from the cascade temporal networks leading to the inhibition stage …

   @article{sarkar2019using,
     pubtype = {published},
     title = {Using network motifs to characterize temporal network evolution leading to diffusion inhibition: S. Sarkar et al.},
     author = {Sarkar, Soumajyoti and Guo, Ruocheng and Shakarian, Paulo},
     journal = {Social Network Analysis and Mining},
     volume = {9},
     number = {1},
     pages = {14},
     year = {2019},
     publisher = {Springer Vienna},
     url = {https://link.springer.com/article/10.1007/s13278-019-0556-z},
   }

2018

1. INITIATOR: Noise-contrastive Estimation for Marked Temporal Point Process

   Ruocheng Guo, Jundong Li, and Huan Liu

   In Proceedings of the 27th International Joint Conference on Artificial Intelligence, 2018

   Abs Bib

   Copious sequential event data has consistently increased in various high-impact domains such as social media and sharing economy. When events start to take place in a sequential fashion, an important question arises:“what type of event will happen at what time in the near future?” To answer the question, a class of mathematical models called the marked temporal point process is often exploited as it can model the timing and properties of events seamlessly in a joint framework. Recently, various recurrent neural network (RNN) models are proposed to enhance the predictive power of mark temporal point process. However, existing marked temporal point models are fundamentally based on the Maximum Likelihood Estimation (MLE) framework for the training, and inevitably suffer from the problem resulted from the intractable likelihood function. Surprisingly, little attention has been paid to address this issue. In this work, we propose INITIATOR-a novel training framework based on noise-contrastive estimation to resolve this problem. Theoretically, we show the exists a strong connection between the proposed INITIATOR and the exact MLE. Experimentally, the efficacy of INITIATOR is demonstrated over the state-of-the-art approaches on several real-world datasets from various areas.

   @inproceedings{guo2018initiator,
     pubtype = {published},
     title = {INITIATOR: Noise-contrastive Estimation for Marked Temporal Point Process},
     author = {Guo, Ruocheng and Li, Jundong and Liu, Huan},
     booktitle = {Proceedings of the 27th International Joint Conference on Artificial Intelligence},
     pages = {2191-2197},
     year = {2018},
     url = {https://www.ijcai.org/Proceedings/2018/0303.pdf},
   }

2. Linked Causal Variational Autoencoder for Inferring Paired Spillover Effects

   Vineeth Rakesh, Ruocheng Guo, Raha Moraffah, and 2 more authors

   In ACM International Conference on Information and Knowledge Management, 2018

   Abs Bib

   Modeling spillover effects from observational data is an important problem in economics, business, and other fields of research. It helps us infer the causality between two seemingly unrelated set of events. For example, if consumer spending in the United States declines, it has spillover effects on economies that depend on the U.S. as their largest export market. In this paper, we aim to infer the causation that results in spillover effects between pairs of entities (or units); we call this effect as paired spillover. To achieve this, we leverage the recent developments in variational inference and deep learning techniques to propose a generative model called Linked Causal Variational Autoencoder (LCVA). Similar to variational autoencoders (VAE), LCVA incorporates an encoder neural network to learn the latent attributes and a decoder network to reconstruct the inputs. However, unlike VAE, LCVA treats the latent attributes …

   @inproceedings{rakesh2018linked,
     pubtype = {published},
     title = {Linked Causal Variational Autoencoder for Inferring Paired Spillover Effects},
     author = {Rakesh, Vineeth and Guo, Ruocheng and Moraffah, Raha and Agarwal, Nitin and Liu, Huan},
     booktitle = {ACM International Conference on Information and Knowledge Management},
     pages = {1679-1682},
     year = {2018},
     url = {https://dl.acm.org/doi/abs/10.1145/3269206.3269267},
   }

3. Detecting pathogenic social media accounts without content or network structure

   Elham Shaabani, Ruocheng Guo, and Paulo Shakarian

   In 2018 1st International Conference on Data Intelligence and Security (ICDIS), 2018

   Abs Bib

   The spread of harmful mis-information in social media is a pressing problem. We refer accounts that have the capability of spreading such information to viral proportions as "Pathogenic Social Media" accounts. These accounts include terrorist supporters accounts, water armies, and fake news writers. We introduce an unsupervised causality-based framework that also leverages label propagation. This approach identifies these users without using network structure, cascade path information, content and user’s information. We show our approach obtains higher precision (0.75) in identifying Pathogenic Social Media accounts in comparison with random (precision of 0.11) and existing bot detection (precision of 0.16) methods.

   @inproceedings{shaabani2018detecting,
     pubtype = {published},
     title = {Detecting pathogenic social media accounts without content or network structure},
     author = {Shaabani, Elham and Guo, Ruocheng and Shakarian, Paulo},
     booktitle = {2018 1st International Conference on Data Intelligence and Security (ICDIS)},
     pages = {57-64},
     year = {2018},
     publisher = {IEEE},
     url = {https://ieeexplore.ieee.org/abstract/document/8367640/},
   }

4. Strongly Hierarchical Factorization Machines and ANOVA Kernel Regression

   Ruocheng Guo, Hamidreza Alvari, and Paulo Shakarian

   In Society for Industrial and Applied Mathematics eBooks, 2018

   DOI Bib

   @inproceedings{guo2018strongly,
     pubtype = {published},
     title = {Strongly Hierarchical Factorization Machines and ANOVA Kernel Regression},
     author = {Guo, Ruocheng and Alvari, Hamidreza and Shakarian, Paulo},
     booktitle = {Society for Industrial and Applied Mathematics eBooks},
     pages = {729--737},
     year = {2018},
     doi = {10.1137/1.9781611975321.82}
   }

2017

1. Understanding and forecasting lifecycle events in information cascades

   Soumajyoti Sarkar, Ruocheng Guo, and Paulo Shakarian

   Social Network Analysis and Mining, 2017

   DOI arXiv Bib

   @article{sarkar2017understanding,
     pubtype = {published},
     title = {Understanding and forecasting lifecycle events in information cascades},
     author = {Sarkar, Soumajyoti and Guo, Ruocheng and Shakarian, Paulo},
     journal = {Social Network Analysis and Mining},
     volume = {7},
     number = {1},
     year = {2017},
     doi = {10.1007/s13278-017-0475-9},
   }

2. Temporal Analysis of Influence to Predict Users’ Adoption in Online Social Networks

   Ericsson Marin, Ruocheng Guo, and Paulo Shakarian

   In Lecture notes in computer science, 2017

   DOI Bib

   @inproceedings{marin2017temporal,
     pubtype = {published},
     title = {Temporal Analysis of Influence to Predict Users’ Adoption in Online Social Networks},
     author = {Marin, Ericsson and Guo, Ruocheng and Shakarian, Paulo},
     booktitle = {Lecture notes in computer science},
     pages = {254--261},
     year = {2017},
     doi = {10.1007/978-3-319-60240-0_31}
   }

2016

1. A comparison of methods for cascade prediction

   Ruocheng Guo and Paulo Shakarian

   In IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2016

   DOI Bib

   @inproceedings{guo2016comparison,
     pubtype = {published},
     title = {A comparison of methods for cascade prediction},
     author = {Guo, Ruocheng and Shakarian, Paulo},
     booktitle = {IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)},
     volume = {12},
     pages = {591--598},
     year = {2016},
     doi = {10.1109/asonam.2016.7752296}
   }

2. An empirical evaluation of social influence metrics

   Nikhil Kumar, Ruocheng Guo, Ashkan Aleali, and 1 more author

   In IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2016

   DOI Bib

   @inproceedings{kumar2016empirical,
     pubtype = {published},
     title = {An empirical evaluation of social influence metrics},
     author = {Kumar, Nikhil and Guo, Ruocheng and Aleali, Ashkan and Shakarian, Paulo},
     booktitle = {IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)},
     pages = {1329--1336},
     year = {2016},
     doi = {10.1109/asonam.2016.7752408}
   }

3. Toward early and order-of-magnitude cascade prediction in social networks

   Ruocheng Guo, Elham Shaabani, Abhinav Bhatnagar, and 1 more author

   Social Network Analysis and Mining, 2016

   DOI Bib

   @article{guo2016toward,
     pubtype = {published},
     title = {Toward early and order-of-magnitude cascade prediction in social networks},
     author = {Guo, Ruocheng and Shaabani, Elham and Bhatnagar, Abhinav and Shakarian, Paulo},
     journal = {Social Network Analysis and Mining},
     volume = {6},
     number = {1},
     year = {2016},
     doi = {10.1007/s13278-016-0372-7}
   }

2015

1. The independent cascade and linear threshold models

   Paulo Shakarian, Abhinav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   In this chapter, we focus on perhaps the two most prevalent diffusion models in computer science—the independent cascade and linear threshold models. We describe different properties of these models and how these properties affect solving problems such as influence maximization and influence spread. We describe approaches to address influence maximization problem in independent cascade model and linear threshold model that rely on the maximization of submodular functions—as well as extensions to these approaches for larger datasets.

   @article{shakarian2015independent,
     pubtype = {published},
     title = {The independent cascade and linear threshold models},
     author = {Shakarian, Paulo and Bhatnagar, Abhinav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {35-48},
     year = {2015},
     url = {https://link.springer.com/chapter/10.1007/978-3-319-23105-1_4},
   }

2. Diffusion in social networks

   Paulo Shakarian, Abhivav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   In recent years, research on diffusion process in social networks has grown in a variety of fields including computer science, physics, and biology. However, often times research in these individual disciplines becomes stove-piped. In this book, we focus on cutting-edge research in social network diffusion bringing together a range of ideas from these disciplines with the goal of creating a single volume that examines these ideas.We sought to cover many of the most important concepts, models, and methods from these areas. We felt by exploring a variety of work from different fields that we could help open the door to more innovative findings in this fascinating area of diffusion in social networks.

   @article{shakarian2015diffusion,
     pubtype = {published},
     title = {Diffusion in social networks},
     author = {Shakarian, Paulo and Bhatnagar, Abhivav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {47-58},
     year = {2015},
     url = {https://link.springer.com/content/pdf/10.1007/978-3-319-23105-1.pdf},
   }

3. Toward Order-of-Magnitude Cascade Prediction

   Ruocheng Guo, Elham Shaabani, Abhinav Bhatnagar, and 1 more author

   In IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2015

   DOI Bib

   @inproceedings{guo2015toward,
     pubtype = {published},
     title = {Toward Order-of-Magnitude Cascade Prediction},
     author = {Guo, Ruocheng and Shaabani, Elham and Bhatnagar, Abhinav and Shakarian, Paulo},
     booktitle = {IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)},
     pages = {1610--1613},
     year = {2015},
     doi = {10.1145/2808797.2809358}
   }

4. The SIR model and identification of spreaders

   Paulo Shakarian, Abhinav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   In this chapter, we review the classic susceptible-infected-recovered (SIR) model for disease spread as applied to a social network. In particular, we look at the problem of identifying nodes that are “spreaders” which cause a large part of the population to become infected under this model. To do so, we survey a variety of nodal measures based on centrality (degree, betweenness, etc.) and other methods (shell decomposition, nearest neighbor analysis, etc.). We then present a set of experiments that illustrate the relation of these nodal measures to spreading under the SIR model.

   @article{shakarian2015model,
     pubtype = {published},
     title = {The SIR model and identification of spreaders},
     author = {Shakarian, Paulo and Bhatnagar, Abhinav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {3-18},
     year = {2015},
     url = {https://link.springer.com/chapter/10.1007/978-3-319-23105-1_2},
   }

5. Evolutionary graph theory

   Paulo Shakarian, Abhinav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   Evolutionary graph theory (EGT), studies the ability of a mutant gene to overtake a finite structured population. In this chapter, we describe the original framework for EGT and the major work that has followed it. Here, we will study the calculation of the “fixation probability”—the probability of a mutant taking over a population and focuses on game-theoretic applications. We look at varying topics such as alternate evolutionary dynamics, time to fixation, special topological cases, and game theoretic results.

   @article{shakarian2015evolutionary,
     pubtype = {published},
     title = {Evolutionary graph theory},
     author = {Shakarian, Paulo and Bhatnagar, Abhinav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {75-91},
     year = {2015},
     url = {https://link.springer.com/chapter/10.1007/978-3-319-23105-1_6},
   }

6. Logic Programming Based Diffusion Models

   Paulo Shakarian, Abhinav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   In this chapter, we first show that the well-known generalized annotated program (GAP) paradigm can be used to express many existing diffusion models that can consider not only the topology of the social network, but attributes of the nodes and edges as well. We then define a class of problems called Social Network Diffusion Optimization Problems (SNDOPs). In this chapter, we show how various diffusion processes can be embedded as GAP’s and then study the algorithmic and complexity issues associated with SDNOP’s. Experimental results are also included.

   @article{shakarian2015logic,
     pubtype = {published},
     title = {Logic Programming Based Diffusion Models},
     author = {Shakarian, Paulo and Bhatnagar, Abhinav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {49-73},
     year = {2015},
     url = {https://link.springer.com/chapter/10.1007/978-3-319-23105-1_5},
   }

7. The tipping model and the minimum seed problem

   Paulo Shakarian, Abhinav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   In a “tipping” model, each node in a social network, representing an individual, adopts a property or behavior if a certain number of his incoming neighbors currently exhibit the same. A key problem, with respect to this model, is to select an initial “seed” set from the network such that the entire network adopts any behavior given to the seed. In this chapter, we investigate the problem of identifying a seed set of minimum size—which is NP-hard. We provide exact and heuristic methods for solving this problem as well as a suite of results to not only demonstrate the utility of these methods, but provide insight into the dynamics of the tipping model also.

   @article{shakarian2015tipping,
     pubtype = {published},
     title = {The tipping model and the minimum seed problem},
     author = {Shakarian, Paulo and Bhatnagar, Abhinav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {19-33},
     year = {2015},
     url = {https://link.springer.com/chapter/10.1007/978-3-319-23105-1_3},
   }

8. Examining Diffusion in the Real World

   Paulo Shakarian, Abhinav Bhatnagar, Ashkan Aleali, and 2 more authors

   Springer International Publishing, 2015

   Abs Bib

   Throughout this book, we studied a variety of diffusion models that are commonly seen in the literature of computer science, physics, and biology. In this chapter, we study diffusion processes from a data-driven perspective—specifiably reviewing the early identification of information cascades that will diffuse through a large portion of the network.

   @article{shakarian2015examining,
     pubtype = {published},
     title = {Examining Diffusion in the Real World},
     author = {Shakarian, Paulo and Bhatnagar, Abhinav and Aleali, Ashkan and Shaabani, Elham and Guo, Ruocheng},
     journal = {Springer International Publishing},
     pages = {93-100},
     year = {2015},
     url = {https://link.springer.com/chapter/10.1007/978-3-319-23105-1_7},
   }

1. T-GINEE: A Tensor-Based Multi-Graph Representation Learning

   Maolin Wang, Ziting Mai, Xuhui Chen, and 9 more authors

   In Forty-third International Conference on Machine Learning

   Abs Bib

   While traditional network analysis focuses on single-layer networks, real-world systems often form multilayer networks with multiple relationship types. However, existing methods typically fail to capture complex inter-layer dependencies by treating layers independently or aggregating them. To address this, we propose T-GINEE (Tensor-Based Generalized Multilayer-graph Estimating Equation), a statistical regularization framework combining tensor-based generalized estimating equations with task-specific loss to model cross-network correlations explicitly. Key innovations include: (1) CP tensor decomposition capturing structural dependencies via shared latent factors; (2) a generalized estimating equation framework modeling inter-layer correlations through working covariance matrices; and (3) a flexible link function accommodating characteristics like sparsity. Our theoretical analysis establishes consistency and asymptotic normality under mild conditions. Extensive experiments on synthetic and real-world datasets validate T-GINEE’s effectiveness for multilayer network analysis. Our code is available in the \textbfsupplementary materials to ensure reproducibility.

   @inproceedings{wangginee,
     pubtype = {published},
     title = {T-GINEE: A Tensor-Based Multi-Graph Representation Learning},
     author = {Wang, Maolin and Mai, Ziting and Chen, Xuhui and Li, Zhiqi and Wei, Tianshuo and Xiao, Yutian and Zhang, Wenlin and Wang, Wanyu and Guo, Ruocheng and Li, Haoxuan and Xu, Zenglin and Zhao, Xiangyu},
     booktitle = {Forty-third International Conference on Machine Learning},
     url = {https://openreview.net/forum?id=hllyaNwbOz},
   }

Preprints (arXiv only)

2026

1. Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Use

   Ruocheng Guo, Kaiwen Dong, Xiang Gao, and 1 more author

   arXiv preprint, 2026

   DOI arXiv Bib

   @article{guo2026learning,
     pubtype = {preprint},
     title = {Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Use},
     author = {Guo, Ruocheng and Dong, Kaiwen and Gao, Xiang and Das, Kamalika},
     journal = {arXiv preprint},
     year = {2026},
     doi = {10.48550/arxiv.2602.20426},
   }

2. When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories

   Avinash Baidya, Xinran Liang, Ruocheng Guo, and 2 more authors

   arXiv preprint, 2026

   DOI arXiv Bib

   @article{baidya2026when,
     pubtype = {preprint},
     title = {When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories},
     author = {Baidya, Avinash and Liang, Xinran and Guo, Ruocheng and Gao, Xiang and Das, Kamalika},
     journal = {arXiv preprint},
     year = {2026},
     doi = {10.48550/arxiv.2606.05414},
   }

3. Observational Policy Ranking for SMB Financial Guidance from Multi-Action Accounting Logs

   Shrutendra Harsola, Vignesh Subrahmaniam, Vikas Raturi, and 7 more authors

   arXiv preprint, 2026

   DOI arXiv Bib

   @article{harsola2026observational,
     pubtype = {preprint},
     title = {Observational Policy Ranking for SMB Financial Guidance from Multi-Action Accounting Logs},
     author = {Harsola, Shrutendra and Subrahmaniam, Vignesh and Raturi, Vikas and Das, Kamalika and Gao, Xiang and Gupta, Kratika and Guo, Ruocheng and Jonnalagedda, Padmaja and Pramod, Ananya and Kumar, Sricharan},
     journal = {arXiv preprint},
     year = {2026},
     doi = {10.48550/arxiv.2608.10050},
   }

4. TeacherGRPO: Closing the Capacity Gap in Reasoning Distillation via Teacher Alignment

   Zhenyu Lei, Zihan Chen, Yaochen Zhu, and 5 more authors

   arXiv preprint, 2026

   DOI arXiv Bib

   @article{lei2026teachergrpo,
     pubtype = {preprint},
     title = {TeacherGRPO: Closing the Capacity Gap in Reasoning Distillation via Teacher Alignment},
     author = {Lei, Zhenyu and Chen, Zihan and Zhu, Yaochen and Feng, Shangbin and Zheng, Zaiyi and Guo, Ruocheng and Dong, Yushun and Li, Jundong},
     journal = {arXiv preprint},
     year = {2026},
     doi = {10.48550/arxiv.2609.33426},
   }

5. Trojanscope: Interpretable backdoor detection for time series forecasting

   Maolin Wang, Pengzhi Zhong, Binhao Wang, and 4 more authors

   2026

   Abs Bib

   Time series forecasting models are increasingly targeted by backdoor attacks, which embed hidden triggers into parameters while preserving accuracy on clean data. Existing defenses from vision and NLP fail to capture the temporal complexity of triggers and offer limited interpretability. We present TrojanScope, a visual analytics framework for reliable and explainable backdoor detection in time series models. TrojanScope integrates (i) block-level substitution to localize contaminated components, (ii) residual filtering with correlation analysis to isolate abnormal temporal channels, and (iii) candidate extraction and optimization to reconstruct the hidden trigger. Our method is supported by theoretical guarantees on detection consistency and recovery accuracy, and it produces intuitive visual evidence of Trojan propagation. Across multiple datasets (including ESA-ADB satellite telemetry) and architectures (PatchTST, N-HiTS), TrojanScope achieves superior accuracy, efficiency, and interpretability compared with state-of-the-art baselines. This work highlights the importance of combining formal guarantees with visual diagnostics for trustworthy and practical backdoor defense in high-stakes time series forecasting.

   @article{wang2026trojanscope,
     pubtype = {preprint},
     title = {Trojanscope: Interpretable backdoor detection for time series forecasting},
     author = {Wang, Maolin and Zhong, Pengzhi and Wang, Binhao and Wei, Tianshuo and BAO, Beining and Guo, Ruocheng and Zhao, Xiangyu},
     year = {2026},
     url = {https://openreview.net/forum?id=yj9cg8rBcL},
   }

6. Beyond Conditional Independence: Root Cause Analysis with Deep Causal Models

   Md Musfiqur Rahman, Kenneth Lee, Ziwei Jiang, and 3 more authors

   arXiv preprint arXiv:2609.36771, 2026

   Abs arXiv Bib

   Root cause analysis (RCA) is a critical problem in many real-world scenarios. RCA enables the identification of faulty or failing mechanisms in a system by comparing anomalous observations with corresponding reference (i.e., regular) observations. However, existing approaches rely either on heuristic methods or on conditional independence tests with a strong unconfoundedness assumption, and thus fail to exploit other complicated distributional constraints in the presence of latent variables. To relax these assumptions, we model the underlying system as a causal model and the anomalous system as a change in the structural functions of the same causal model. Specifically, to handle unobserved confounders, we establish an implicit connection between distributional constraint testing and root cause analysis. To adapt our approach to data generated from arbitrary causal models, we employ the deep causal model (DCM) framework, in which we design the causal model using neural networks. Finally, we illustrate how our method, RCA-DCM, can utilize different levels of partial graphical knowledge to perform RCA. We evaluate RCA-DCM against state-of-the-art baselines on simulated datasets, a physics-based causal chamber and two micro-service applications. RCA-DCM improves top-1 accuracy over the strongest baseline on both Sock Shop (0.880 vs. 0.752) and Online Boutique (0.776 vs. 0.712), and when the true root cause in the causal chamber is unobserved and acts as a latent confounder, it recovers the exact root-cause set more often than any competing method (perfect recovery rate (PRR) 0.846 vs. 0.731).

   @article{rahman2026beyond,
     pubtype = {preprint},
     title = {Beyond Conditional Independence: Root Cause Analysis with Deep Causal Models},
     author = {Rahman, Md Musfiqur and Lee, Kenneth and Jiang, Ziwei and Jonnalagedda, Padmaja and Guo, Ruocheng and Kocaoglu, Murat},
     journal = {arXiv preprint arXiv:2609.36771},
     year = {2026},
     url = {https://arxiv.org/abs/2609.36771},
   }

2025

1. Learning Counterfactual Outcomes Under Rank Preservation

   Peng Wu, Haoxuan Li, Chunyuan Zheng, and 5 more authors

   arXiv preprint arXiv:2502.06398, 2025

   arXiv Bib

   @article{wu2025learning,
     pubtype = {preprint},
     title = {Learning Counterfactual Outcomes Under Rank Preservation},
     author = {Wu, Peng and Li, Haoxuan and Zheng, Chunyuan and Zeng, Yan and Chen, Jiawei and Liu, Yang and Guo, Ruocheng and Zhang, Kun},
     journal = {arXiv preprint arXiv:2502.06398},
     year = {2025}
   }

2. Identification and Estimation of Long-Term Treatment Effects with Monotone Missing

   Qinwei Yang, Ruocheng Guo, Shasha Han, and 1 more author

   arXiv preprint arXiv:2504.19527, 2025

   arXiv Bib

   @article{yang2025identification,
     pubtype = {preprint},
     title = {Identification and Estimation of Long-Term Treatment Effects with Monotone Missing},
     author = {Yang, Qinwei and Guo, Ruocheng and Han, Shasha and Wu, Peng},
     journal = {arXiv preprint arXiv:2504.19527},
     year = {2025}
   }

3. CuFun: Revolutionizing Temporal Point Process Modeling with Cumulative Distribution Functions

   Maolin Wang, Ruocheng Guo, and Xiangyu Zhao

   arXiv preprint, 2025

   Abs arXiv Bib

   Temporal Point Processes (TPPs) are crucial for modeling event sequences in diverse domains such as social networking and ecommerce. These processes utilize interactions and transactions to identify patterns and predict future trends. Despite their utility, the complexity of the data often hinders the accuracy of these predictions. Recent advancements have merged Neural Networks with TPPs, creating sophisticated models capable of handling complex, nonlinear temporal data. However, challenges persist regarding the flexibility in modeling intensity functions, the computational overhead of integral calculations, and the capture of long-range dependencies. We introduce the CuFun model, a novel TPP approach employing a monotonic neural network to represent the Cumulative Distribution Function (CDF). This model enhances adaptability and accuracy by incorporating past events as scaling factors, simplifies log-likelihood calculations, and extends the model’s applicability beyond conventional density function constraints. CuFun excels in capturing long-range temporal patterns, making it particularly effective for precise forecasting in real-world applications involving extended temporal sequences. Our proposed CDF-based TPP model integrates past event data to improve future predictions.

   @article{wang2025cufun,
     pubtype = {preprint},
     journal = {arXiv preprint},
     title = {CuFun: Revolutionizing Temporal Point Process Modeling with Cumulative Distribution Functions},
     author = {Wang, Maolin and Guo, Ruocheng and Zhao, Xiangyu},
     year = {2025},
     url = {https://kdd.org/kdd2025/wp-content/uploads/2025/07/paper_11.pdf},
   }

2024

1. Adversarial Curriculum Graph Contrastive Learning with Pair-wise Augmentation

   Xinjian Zhao, Liang Zhang, Yang Liu, and 2 more authors

   arXiv preprint, 2024

   DOI arXiv Bib

   @article{zhao2024adversarial,
     pubtype = {preprint},
     title = {Adversarial Curriculum Graph Contrastive Learning with Pair-wise Augmentation},
     author = {Zhao, Xinjian and Zhang, Liang and Liu, Yang and Guo, Ruocheng and Zhao, Xiangyu},
     journal = {arXiv preprint},
     year = {2024},
     doi = {10.48550/arxiv.2402.10468},
   }

2023

1. Trustworthy llms: a survey and guideline for evaluating large language models’ alignment

   Yang Liu, Yuanshun Yao, Jean-Francois Ton, and 6 more authors

   arXiv preprint arXiv:2308.05374, 2023

   arXiv Bib

   @article{liu2023trustworthy,
     pubtype = {preprint},
     title = {Trustworthy llms: a survey and guideline for evaluating large language models' alignment},
     author = {Liu, Yang and Yao, Yuanshun and Ton, Jean-Francois and Zhang, Xiaoying and Guo, Ruocheng and Cheng, Hao and Klochkov, Yegor and Taufiq, Muhammad Faaiz and Li, Hang},
     journal = {arXiv preprint arXiv:2308.05374},
     year = {2023}
   }

2. Inference-time Stochastic Ranking with Risk Control

   Ruocheng Guo, Jean-François Ton, and Yang Liu

   arXiv preprint arXiv:2306.07188, 2023

   arXiv Bib

   @article{guo2023inference,
     pubtype = {preprint},
     title = {Inference-time Stochastic Ranking with Risk Control},
     author = {Guo, Ruocheng and Ton, Jean-Fran{\c{c}}ois and Liu, Yang},
     journal = {arXiv preprint arXiv:2306.07188},
     year = {2023}
   }

3. Deep Concept Removal

   Yegor Klochkov, Jean-François Ton, Ruocheng Guo, and 2 more authors

   arXiv preprint, 2023

   DOI arXiv Bib

   @article{klochkov2023deep,
     pubtype = {preprint},
     title = {Deep Concept Removal},
     author = {Klochkov, Yegor and Ton, Jean-François and Guo, Ruocheng and Liu, Yang and Li, Hang},
     journal = {arXiv preprint},
     year = {2023},
     doi = {10.48550/arxiv.2310.05755},
   }

2021

1. Out-of-distribution prediction with invariant risk minimization: The limitation and an effective fix

   Ruocheng Guo, Pengchuan Zhang, Hao Liu, and 1 more author

   arXiv preprint arXiv:2101.07732, 2021

   Abs arXiv Bib

   This work considers the out-of-distribution (OOD) prediction problem where (1) the training data are from multiple domains and (2) the test domain is unseen in the training. DNNs fail in OOD prediction because they are prone to pick up spurious correlations. Recently, Invariant Risk Minimization (IRM) is proposed to address this issue. Its effectiveness has been demonstrated in the colored MNIST experiment. Nevertheless, we find that the performance of IRM can be dramatically degraded under \emphstrong spuriousness – when the spurious correlation between the spurious features and the class label is strong due to the strong causal influence of their common cause, the domain label, on both of them (see Fig. 1). In this work, we try to answer the questions: why does IRM fail in the aforementioned setting? Why does IRM work for the original colored MNIST dataset? How can we fix this problem of IRM? Then, we propose a simple and effective approach to fix the problem of IRM. We combine IRM with conditional distribution matching to avoid a specific type of spurious correlation under strong spuriousness. Empirically, we design a series of semi synthetic datasets – the colored MNIST plus, which exposes the problems of IRM and demonstrates the efficacy of the proposed method.

   @article{guo2021distribution,
     pubtype = {preprint},
     title = {Out-of-distribution prediction with invariant risk minimization: The limitation and an effective fix},
     author = {Guo, Ruocheng and Zhang, Pengchuan and Liu, Hao and Kiciman, Emre},
     journal = {arXiv preprint arXiv:2101.07732},
     year = {2021},
     url = {https://arxiv.org/abs/2101.07732},
   }

2. Learning Causality with Networked Observational Data

   Ruocheng Guo

   2021

   Abs Bib

   This dissertation considers the question of how convenient access to copious networked observational data impacts our ability to learn causal knowledge. It investigates in what ways learning causality from such data is different from–or the same as–the traditional causal inference which often deals with small scale iid data collected from randomized controlled trials? For example, how can we exploit network information for a series of tasks in the area of learning causality? To answer this question, the dissertation is written toward developing a suite of novel causal learning algorithms that offer actionable insights for a series of causal inference tasks with networked observational data. The work aims to benefit real-world decision-making across a variety of highly influential applications. In the first part of this dissertation, it investigates the task of inferring individual-level causal effects from networked observational …

   @article{guo2021learning,
     pubtype = {preprint},
     title = {Learning Causality with Networked Observational Data},
     author = {Guo, Ruocheng},
     year = {2021},
     url = {https://search.proquest.com/openview/5658f223b1e235f0bab5d505b298169e/1?pq-origsite=gscholar&cbl=18750&diss=y},
   }

2019

1. Protecting user privacy: An approach for untraceable web browsing history and unambiguous user profiles

   Ghazaleh Beigi, Ruocheng Guo, Alexander Nou, and 2 more authors

   2019

   Abs Bib

   The overturning of the Internet Privacy Rules by the Federal Communications Commissions (FCC) in late March 2017 allows Internet Service Providers (ISPs) to collect, share and sell their customers’ Web browsing data without their consent. With third-party trackers embedded on Web pages, this new rule has put user privacy under more risk. The need arises for users on their own to protect their Web browsing history from any potential adversaries. Although some available solutions such as Tor, VPN, and HTTPS can help users conceal their online activities, their use can also significantly hamper personalized online services, i.e., degraded utility. In this paper, we design an effective Web browsing history anonymization scheme, PBooster, aiming to protect users’ privacy while retaining the utility of their Web browsing history. The proposed model pollutes users’ Web browsing history by automatically inferring how …

   @article{beigi2019protecting,
     pubtype = {preprint},
     title = {Protecting user privacy: An approach for untraceable web browsing history and unambiguous user profiles},
     author = {Beigi, Ghazaleh and Guo, Ruocheng and Nou, Alexander and Zhang, Yanchao and Liu, Huan},
     pages = {213-221},
     year = {2019},
     url = {https://dl.acm.org/doi/abs/10.1145/3289600.3291026},
   }

2. Toward Controlling Hidden Confounding for Causal Effect Estimation with Networked Observational Data

   Ruocheng Guo, Jundong Li, and Huan Liu

   arXiv preprint, 2019

   arXiv Bib

   @article{guo2019toward,
     pubtype = {preprint},
     title = {Toward Controlling Hidden Confounding for Causal Effect Estimation with Networked Observational Data},
     author = {Guo, Ruocheng and Li, Jundong and Liu, Huan},
     journal = {arXiv preprint},
     year = {2019},
   }

1. Collaborative Dual-Size Large Language Models with Dual-Stage Deferral Risk Control

   Maolin Wang, Binhao Wang, Yingyi Zhang, and 7 more authors

   Abs Bib

   Large Language Models (LLMs) have demonstrated remarkable capabilities, yet ensuring their safe deployment remains challenging. Existing safety mechanisms, while effective against malicious inputs, often degrade performance on benign queries due to over-conservative strategies. We propose the \textbfDual-size LLM collaborative framework with \textbfDual-stage deferral risk contro\textbfL (\textbfDDL), which integrates lightweight and heavyweight models with calibrated deferral mechanisms. Our approach formalizes the safety–efficiency trade-off as a constrained optimization problem that jointly considers prediction accuracy, computational cost, and safety risk. We provide theoretical guarantees showing that our mechanism achieves distribution-free risk control while minimizing unnecessary heavyweight computation. Extensive experiments on three datasets demonstrate that DDL effectively balances safety and efficiency, achieving performance and safety metrics comparable to state-of-the-art safety-aligned models while reducing average inference time by more than 65\%.

   @article{wangcollaborative,
     pubtype = {preprint},
     title = {Collaborative Dual-Size Large Language Models with Dual-Stage Deferral Risk Control},
     author = {Wang, Maolin and Wang, Binhao and Zhang, Yingyi and Liu, Rungen and Wang, Wanyu and Wei, Xuetao and Zou, Lixin and Yin, Hongzhi and Guo, Ruocheng and Zhao, Xiangyu},
     url = {https://openreview.net/forum?id=KHIeyfEqZE},
   }