[{"id":"deltagate","title":"Zero-Observation User Reactivation with Gap-Driven Dimensional Gating","year":2026,"venue":"Accepted to the 20th ACM Conference on Recommender Systems (RecSys '26)","venue_short":"RecSys","authors":"J Ding, T Liu, F Liu, H Qin, T Wu","topic":"recommender-systems","level":"a","selected":true,"image":"/assets/images/DeltaGate-Resurrecting-Ghost-Users.jpg","selected_label":"RecSys 2026","selected_summary":"DeltaGate recalibrates frozen sequential recommendation representations for long-gap returning users.","code_url":"https://github.com/jdding/DeltaGate","code_label":"Code","digest_url":"/deltagate/","digest_label":"Digest","digest_note":"A RecSys 2026 accepted paper defining zero-observation user reactivation and introducing DeltaGate, a frozen-backbone dimensional gating plugin for long-gap returning users.","digest_blocks":[{"title":"Problem","text":"Returning users may have months or years without observed behavior, so a model's last user representation can become unreliable even when pre-gap history exists."},{"title":"Method","text":"DeltaGate keeps the sequential recommendation backbone frozen and learns a dimension-wise gate that routes each representation dimension between personalized history and a learned global prior."},{"title":"Findings","text":"Under the Gap-Synthesize Protocol on Amazon datasets, standard sequential recommenders decline as inactivity gaps widen; in the >365d Video Games bucket, DG-BERT4Rec reaches 0.046 Hit@10 versus 0.025 for BERT4Rec."},{"title":"Deployment boundary","text":"End-to-end retraining can reach higher absolute accuracy, while the frozen DeltaGate plugin preserves zero backbone drift, uses 66K trainable parameters, and keeps the adaptation behavior inspectable."}]},{"id":"skillresolve-bench","title":"SkillResolve-Bench: Measuring and Resolving Same-Capability Ambiguity in Agent Skill Retrieval","year":2026,"venue":"arXiv preprint arXiv:2606.10388","venue_short":"arXiv","authors":"J Ding","topic":"llm-agents","level":"c","selected":false,"paper_url":"https://arxiv.org/abs/2606.10388","paper_label":"arXiv","digest_url":"/skillresolve-bench/","digest_label":"Digest","digest_note":"A benchmark-focused study of same-capability ambiguity in agent skill retrieval, where similar-looking skills must be distinguished by execution constraints and task fit."},{"id":"sidinspector","title":"SIDInspector: A Mapping-First Diagnostic Resource for Semantic-ID Tokenizers","year":2026,"venue":"arXiv preprint arXiv:2606.10375","venue_short":"arXiv","authors":"J Ding, H Chang, H Qin, T Liu","topic":"llm-agents","level":"c","selected":false,"paper_url":"https://arxiv.org/abs/2606.10375","paper_label":"arXiv","digest_url":"/sidinspector/","digest_label":"Digest","digest_note":"A diagnostic resource for inspecting Semantic-ID tokenizers through their mapping behavior before downstream recommendation training."},{"id":"hpgr","title":"Beyond the Flat Sequence: Hierarchical and Preference-Aware Generative Recommendations","year":2026,"venue":"Proceedings of the ACM Web Conference 2026","venue_short":"WWW","authors":"Z Chen, H Chang, T Liu, C Zhou, Y Cao, J Ding, M Liu, B Qin","topic":"recommender-systems","level":"a","selected":true,"image":"/assets/images/Beyond-the-Flat-Sequence.webp","selected_label":"WWW 2026","selected_summary":"Hierarchical and preference-aware generative recommendations move beyond flat user histories and expose structured intent over time.","digest_url":"/hpgr/","digest_label":"Digest","doi_url":"https://doi.org/10.1145/3774904.3792790","digest_note":"A generative recommendation paper that studies hierarchical structure and preference-aware modeling beyond flat user-history sequences."},{"id":"invariant-feature-learning","title":"Invariant feature learning for counterfactual watch-time prediction in video recommendation","year":2026,"venue":"Proceedings of the AAAI Conference on Artificial Intelligence","venue_short":"AAAI","authors":"C Jin, Y Ren, H Ma, Y Xia, Y Guan, H Zhang, J Ding, J Guan, S Zhou","topic":"recommender-systems","level":"a","selected":true,"image":"/assets/images/Invariant-Feature-Learning.webp","selected_label":"AAAI 2026","selected_summary":"Invariant feature learning separates duration effects from true user preference in short-video recommendation.","digest_url":"/aaai-difl/","digest_label":"Digest","doi_url":"https://doi.org/10.1609/aaai.v40i17.38518","digest_note":"A video recommendation paper on counterfactual watch-time prediction and feature learning under duration-related bias."},{"id":"rpe4rec","title":"RPE4Rec: Enhancing Dynamic Node Retrieval with Efficient Relative Position Encoding for Recommendation Systems","year":2026,"venue":"Proceedings of the Nineteenth ACM International Conference on Web Search and Data Mining","venue_short":"WSDM","authors":"K Cheng, H Chang, P Wang, L Gu, J Ding, Y Cao, J Ye, B Du","topic":"recommender-systems","level":"b","selected":true,"image":"/assets/images/RPE4Rec.webp","selected_label":"WSDM 2026","selected_summary":"Efficient relative position encoding for dynamic node retrieval in recommendation systems.","paper_url":"https://dl.acm.org/doi/epdf/10.1145/3773966.3778006","paper_label":"PDF","digest_url":"/rpe4rec/","digest_label":"Digest","doi_url":"https://doi.org/10.1145/3773966.3778006","digest_note":"A recommendation retrieval paper on relative position encoding for dynamic node retrieval under changing user and item behavior."},{"id":"dygraph","title":"Towards Practical Large-scale Dynamical Heterogeneous Graph Embedding: Cold-start Resilient Recommendation","year":2025,"venue":"arXiv preprint arXiv:2512.13120","venue_short":"arXiv","authors":"M Long, J Liu, Y Li, H Xiong, J Yan, K Wang, Y Cao, J Ding","topic":"recommender-systems","level":"c","selected":false,"paper_url":"https://arxiv.org/abs/2512.13120","paper_label":"arXiv","digest_url":"/dygraph/","digest_label":"Digest","doi_url":"https://doi.org/10.48550/arxiv.2512.13120","digest_note":"A dynamic heterogeneous graph embedding study focused on recommendation settings where new users, items, and relations appear over time."},{"id":"bis-nl2sql","title":"BIS: NL2SQL Service Evaluation Benchmark for Business Intelligence Scenarios","year":2025,"venue":"Lecture Notes in Computer Science, 15405, 357-372","venue_short":"LNCS","authors":"B Caglayan, M Wang, JD Kelleher, S Fei, G Tong, J Ding, P Zhang","topic":"llm-agents","level":"b","selected":true,"image":"/assets/images/BIS-NL2SQL.webp","selected_label":"LNCS 2025","selected_summary":"A benchmark for evaluating service-oriented NL2SQL systems in business intelligence scenarios.","paper_url":"https://doi.org/10.1007/978-981-96-0808-9_27","paper_label":"DOI","digest_url":"/bis-nl2sql/","digest_label":"Digest","doi_url":"https://doi.org/10.1007/978-981-96-0808-9_27","digest_note":"A benchmark and evaluation study for NL2SQL services in business intelligence scenarios, connecting language interfaces with service-oriented data systems."},{"id":"unified-low-rank-compression","title":"Unified Low-rank Compression Framework for Click-through Rate Prediction","year":2024,"venue":"Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","venue_short":"KDD","authors":"H Yu, M Fu, J Ding, Y Zhou, J Wu","topic":"recommender-systems","level":"a","selected":true,"image":"/assets/images/Unified-Low-rank-Compression.webp","selected_label":"KDD 2024","selected_summary":"A unified compression framework for large CTR models and resource-constrained deployment.","paper_url":"https://dl.acm.org/doi/epdf/10.1145/3637528.3671520","paper_label":"PDF","digest_url":"/kdd-ctr/","digest_label":"Digest","doi_url":"https://doi.org/10.1145/3637528.3671520","digest_note":"A CTR prediction paper on low-rank compression for reducing model cost while preserving practical ranking behavior."},{"id":"pd-serve","title":"P/D-Serve: Serving Disaggregated Large Language Model at Scale","year":2024,"venue":"arXiv preprint arXiv:2408.08147","venue_short":"arXiv","authors":"Y Jin, T Wang, H Lin, M Song, P Li, Y Ma, Y Shan, Z Yuan, C Li, Y Sun, et al.","topic":"llm-agents","level":"c","selected":false,"paper_url":"https://arxiv.org/pdf/2408.08147","paper_label":"PDF","digest_url":"/pd-serve/","digest_label":"Digest","doi_url":"https://doi.org/10.48550/arxiv.2408.08147","digest_note":"A systems paper on disaggregated large language model serving, separating serving stages to support large-scale inference."},{"id":"acl-topic","title":"Neural Topic Modeling Based on Cycle Adversarial Training and Contrastive Learning","year":2023,"venue":"Findings of the Association for Computational Linguistics: ACL 2023","venue_short":"ACL Findings","authors":"B Wang, L Zhang, D Zhou, Y Cao, J Ding","topic":"data-mining","level":"a","selected":false,"paper_url":"https://aclanthology.org/2023.findings-acl.616.pdf","paper_label":"PDF","digest_url":"/acl-topic/","digest_label":"Digest","doi_url":"https://doi.org/10.18653/v1/2023.findings-acl.616","digest_note":"A neural topic modeling paper that combines cycle adversarial training and contrastive learning to improve topic representations."},{"id":"continual-gcn","title":"Continual Graph Convolutional Network for Text Classification","year":2023,"venue":"Proceedings of the AAAI Conference on Artificial Intelligence","venue_short":"AAAI","authors":"T Wu, Q Liu, Y Cao, Y Huang, XM Wu, J Ding","topic":"data-mining","level":"a","selected":false,"paper_url":"https://ojs.aaai.org/index.php/AAAI/article/view/26611/26383","paper_label":"PDF","digest_url":"/continual-gcn/","digest_label":"Digest","doi_url":"https://doi.org/10.1609/aaai.v37i11.26611","digest_note":"A graph-learning paper for continual text classification, studying how graph convolutional models adapt as text data evolves."},{"id":"naacl-epida","title":"EPiDA: An Easy Plug-in Data Augmentation Framework for High Performance Text Classification","year":2022,"venue":"Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics","venue_short":"NAACL","authors":"M Zhao, L Zhang, Y Xu, J Ding, J Guan, S Zhou","topic":"data-mining","level":"b","selected":false,"paper_url":"https://aclanthology.org/2022.naacl-main.349","paper_label":"PDF","digest_url":"/naacl-epida/","digest_label":"Digest","doi_url":"https://doi.org/10.18653/v1/2022.naacl-main.349","digest_note":"A text-classification paper introducing an easy plug-in data augmentation framework for low-resource and high-performance settings."},{"id":"neurips-dpssl","title":"DP-SSL: Towards Robust Semi-Supervised Learning with a Few Labeled Samples","year":2021,"venue":"Advances in Neural Information Processing Systems","venue_short":"NeurIPS","authors":"Y Xu, J Ding, L Zhang, S Zhou","topic":"data-mining","level":"a","selected":false,"paper_url":"https://arxiv.org/pdf/2110.13740","paper_label":"PDF","digest_url":"/neurips-dpssl/","digest_label":"Digest","doi_url":"https://doi.org/10.5555/3540261.3541477","digest_note":"A robust semi-supervised learning paper that combines data programming signals with learning from few labeled examples."},{"id":"emnlp-keygraph","title":"Weakly-Supervised Text Classification Based on Keyword Graph","year":2021,"venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","venue_short":"EMNLP","authors":"L Zhang, J Ding, Y Xu, Y Liu, S Zhou","topic":"data-mining","level":"b","selected":false,"paper_url":"https://aclanthology.org/2021.emnlp-main.222.pdf","paper_label":"PDF","digest_url":"/emnlp-keygraph/","digest_label":"Digest","doi_url":"https://doi.org/10.18653/v1/2021.emnlp-main.222","digest_note":"A weakly supervised text classification paper that uses keyword graphs to propagate limited supervision into useful classification signals."},{"id":"ai-assistance-live-streaming","title":"Effectiveness of AI Assistance in Live-Streaming: A Randomized Field Experiment","year":2021,"venue":"ICIS 2021 Proceedings","venue_short":"ICIS","authors":"L Wang, Y He, J Ding, N Huang, Y Hong, X Guo, D Liu, G Chen","topic":"data-mining","level":"a","selected":false,"paper_url":"https://aisel.aisnet.org/icis2021/hci_robot/hci_robot/3/","paper_label":"PDF","digest_url":"/ai-assistance-live-streaming/","digest_label":"Digest","digest_note":"A field-experiment study of AI assistance in live streaming, focusing on how algorithmic support changes creator behavior and platform outcomes."},{"id":"sales-data-live-streaming","title":"The Sales Data Sells: Effects of Real-Time Sales Analytics on Live Streaming Selling","year":2021,"venue":"ICIS 2021 Proceedings","venue_short":"ICIS","authors":"Y He, L Wang, N Huang, Y Hong, J Ding, Y Sun, Y Liu","topic":"data-mining","level":"a","selected":false,"paper_url":"https://aisel.aisnet.org/icis2021/data_analytics/data_analytics/2/","paper_label":"PDF","digest_url":"/sales-data-live-streaming/","digest_label":"Digest","digest_note":"A live-streaming field study on how real-time sales analytics affects seller behavior and selling performance."},{"id":"large-scale-mirna-clustering","title":"Automatically Clustering Large-Scale miRNA Sequences: Methods and Experiments","year":2012,"venue":"BMC Genomics","venue_short":"BMC Genomics","authors":"L Wan, J Ding, T Jin, J Guan, S Zhou","topic":"data-mining","topic_page":false,"level":"b","selected":false,"paper_url":"https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-13-S8-S15","paper_label":"PDF","digest_url":"/large-scale-mirna-clustering/","digest_label":"Digest","doi_url":"https://doi.org/10.1186/1471-2164-13-s8-s15","digest_note":"An early data-mining paper on clustering large-scale miRNA sequences while balancing scale, similarity, and biological interpretation."},{"id":"finding-microrna-targets-plants","title":"Finding MicroRNA Targets in Plants: Current Status and Perspectives","year":2012,"venue":"Genomics, Proteomics and Bioinformatics","venue_short":"GPB","authors":"J Ding, S Zhou, J Guan","topic":"data-mining","topic_page":false,"level":"a","selected":false,"paper_url":"https://academic.oup.com/gpb/article/10/5/264/7221796","paper_label":"PDF","digest_url":"/finding-microrna-targets-plants/","digest_label":"Digest","doi_url":"https://doi.org/10.1016/j.gpb.2012.09.003","digest_note":"A review paper on computational and experimental approaches for finding microRNA targets in plants."},{"id":"genome-wide-mirna-target-interactions","title":"Genome-Wide Search for miRNA-target Interactions in Arabidopsis Thaliana with an Integrated Approach","year":2012,"venue":"BMC Genomics","venue_short":"BMC Genomics","authors":"J Ding, D Li, U Ohler, J Guan, S Zhou","topic":"data-mining","topic_page":false,"level":"b","selected":false,"paper_url":"https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-13-S3","paper_label":"PDF","digest_url":"/genome-wide-mirna-target-interactions/","digest_label":"Digest","doi_url":"https://doi.org/10.1186/1471-2164-13-s3-s3","digest_note":"A genome-scale bioinformatics paper on identifying miRNA-target interactions in Arabidopsis thaliana through integrated computational evidence."},{"id":"mirfam","title":"miRFam: An Effective Automatic miRNA Classification Method Based on n-Grams and a Multiclass SVM","year":2011,"venue":"BMC Bioinformatics","venue_short":"BMC Bioinformatics","authors":"J Ding, S Zhou, J Guan","topic":"data-mining","topic_page":false,"level":"a","selected":false,"paper_url":"https://doi.org/10.1186/1471-2105-12-216","paper_label":"DOI","digest_url":"/mirfam/","digest_label":"Digest","doi_url":"https://doi.org/10.1186/1471-2105-12-216","digest_note":"An early sequence-mining paper that represents miRNA sequences with n-gram features for family classification."},{"id":"imirtp","title":"imiRTP: An Integrated Method to Identifying miRNA-target Interactions in Arabidopsis Thaliana","year":2011,"venue":"2011 IEEE International Conference on Bioinformatics and Biomedicine","venue_short":"BIBM","authors":"J Ding, S Yu, U Ohler, J Guan, S Zhou","topic":"data-mining","topic_page":false,"level":"b","selected":false,"paper_url":"https://ieeexplore.ieee.org/document/6120415/","paper_label":"PDF","digest_url":"/imirtp/","digest_label":"Digest","doi_url":"https://doi.org/10.1109/bibm.2011.13","digest_note":"An integrated method for identifying miRNA-target interactions in Arabidopsis thaliana by combining multiple computational signals."},{"id":"mirensvm","title":"MiRenSVM: Towards Better Prediction of microRNA Precursors Using an Ensemble SVM Classifier with Multi-Loop Features","year":2010,"venue":"BMC Bioinformatics","venue_short":"BMC Bioinformatics","authors":"J Ding, S Zhou, J Guan","topic":"data-mining","topic_page":false,"level":"a","selected":false,"paper_url":"https://doi.org/10.1186/1471-2105-11-S11-S11","paper_label":"DOI","digest_url":"/mirensvm/","digest_label":"Digest","doi_url":"https://doi.org/10.1186/1471-2105-11-s11-s11","digest_note":"An early machine-learning paper for microRNA precursor prediction using ensemble SVM models and multi-loop structural features."}]