Skip to main content Report an accessibility problem ASU Home My ASU Colleges and Schools Sign In
Arizona State University logo
Ira A. Fulton Schools of Engineering
Yanjie Fu
  • Home
  • Research
  • Publications
  • Teaching
  • Team

Publications

  1. Home
  2. Publications

Google Scholar

https://scholar.google.com/citations?hl=en&user=OSArex4AAAAJ&view_op=list_works

Recent Selected Publications

  1. [KDD26] Geometry-First Generative Spatial Single-Cell Reconstruction
  2. [ACM AI Letters] Generative AI Meets Future Cities: Towards an Era of Autonomous Urban Intelligence
  3. [ICML26] Not All Answers Are Contextually Persuadable: Inference Dynamics in Large Language Models under Contextual Influence
  4. [ICLR26] MMTok: Multimodal Coverage Maximization for Efficient Inference of VLMs
  5. [AAAI26] DGTF: Cross-Domain Decentralized Graph Learning with Topology-Aware Knowledge Fusion
  6. [AAAI26] Brownian Bridge Augmented Surrogate Simulation and Injection Planning for Geological CO$_2$ Storage
  7. [AAAI26] Efficient Post-Training Refinement of Latent Reasoning in Large Language Models
  8. [WWW26] Learning to Route: A Rule-Driven Agent Framework for Hybrid-Source Retrieval-Augmented Generation
  9. [EACL26] Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement
  10. [EACL26] Blind Spot Navigation in Large Language Model Reasoning with Thought Space Explorer
  11. [SIGKDD Explorations] Urban Planning in the Age of Agentic AI: Emerging Paradigms and Prospects
  12. Autonomous Data Agents: A New Opportunity for Smart Data
  13. [NeurIPS25] Sculpting Features from Noise: Reward-Guided Hierarchical Diffusion for Task-Optimal Feature Transformation
  14. [TKDD] Topology-aware Reinforcement Feature Space Reconstruction for Graph Data
  15. [TKDE] A Comprehensive Survey on Data Augmentation
  16. [EMNLP25] Weaver: Interweaving SQL and LLM for Table Reasoning
  17. [EMNLP25] ISACL: Internal State Analyzer for Copyrighted Training Data Leakage
  18. [CIKM25] Adaptive Context-Infused Performance Evaluator for Iterative Feature Space Optimization
  19. [KDD25] Continuous Optimization for Feature Selection with Permutation-Invariant Embedding and Policy-Guided Search
  20. [KDD25] Biological Pathway Guided Gene Selection Through Collaborative Reinforcement Learning
  21. [KDD25] IN-Flow: Instance Normalization Flow for Non-stationary Time Series Forecasting
  22. [IJCAI25] Unsupervised Feature Transformation via In-context Generation, Generator-critic LLM Agents, and Duet-play Teaming
  23. A Survey on Data-Centric AI: Tabular Learning from Reinforcement Learning and Generative AI Perspective
  24. [ICDE25] FastFT: Accelerating Reinforced Feature Transformation via Advanced Exploration Strategies
  25. [TIST] Neuro-symbolic embedding for short and effective feature selection via autoregressive generation.
  26. [TKDD] Feature selection as deep sequential generative learning
  27. [NACCL25] MixLLM: Dynamic Routing in Mixed Large Language Models
  28. [AAAI 25] Evolutionary large language model for automated feature transformation
  29. [KDD24] Unsupervised generative feature transformation via graph contrastive pre-training and multi-objective fine-tuning
  30. [KDD24] Irregular Traffic Time Series Forecasting Based on Asynchronous Spatio-Temporal Graph Convolutional Networks

Research Projects

AI4Data-LLM & Agents

Our studies starts with an Evolutionary Large Language Model (LLM) for generating initial feature transformations, then advances to an unsupervised, agentic approach using Generator-Critic LLM agents and Duet-Play Teaming for transformation and criticism, culminating in a unified Agentic Feature Augmentation process that integrates selection and generation through teaming, planning, and memories.

  • Autonomous Data Agents: A New Opportunity for Smart Data (arxiv)
  • Evolutionary large language model for automated feature transformation (AAAI25)
  • Unsupervised feature transformation via in-context generation, generator-critic llm agents, and duet-play teaming (IJCAI25)
  • Agentic feature augmentation: Unifying selection and generation with teaming, planning, and memories (arxiv)

AI4Data-GenAI

We explore controllable generative AI systems for feature selection, generation, and transformation, leveraging techniques such as embedding space optimization, categorical hashing, autoregressive modeling, graph contrastive pre-training, and neuro-symbolic embedding. Together, they aim to build a unified paradigm for efficient, interpretable, and privacy-preserving feature engineering in machine learning.

  • Beyond discrete selection: Continuous embedding space optimization for generative feature selection (ICDM23)
  • Self-optimizing feature generation via categorical hashing representation and hierarchical reinforcement crossing (ICDM23)
  • Reinforcement-enhanced autoregressive feature transformation: Gradient-steered search in continuous space for postfix expressions (NIPS23)
  • Unsupervised generative feature transformation via graph contrastive pre-training and multi-objective fine-tuning (KDD24)
  • Feature selection as deep sequential generative learning (ACM TKDD)
  • Neuro-symbolic embedding for short and effective feature selection via autoregressive generation (ACM TIST)
  • Privacy-preserving data reprogramming (NPJ Artificial Intelligence)
  • DELTA: Variational Disentangled Learning for Privacy-Preserving Data Reprogramming (ICDM25)

AI4Data-RL

We explore feature selection and generation using reinforcement learning (RL), proposing methods like interactive RL with decision trees, multi-agent RL for subspace exploration, and diversity-aware RL to optimize feature subsets. These approaches automate and enhance feature selection by leveraging single or multi-agent strategies, hierarchical RL, and categorical hashing, improving model accuracy, scalability, and interpretability across high-dimensional datasets.

  • Interactive reinforcement learning for feature selection with decision tree in the loop (TKDE21)
  • Automated feature selection: A reinforcement learning perspective (TKDE21)
  • Automating feature subspace exploration via multi-agent reinforcement learning (SIGKDD19)
  • Autofs: Automated feature selection via diversity-aware interactive reinforcement learning (ICDM20)
  • Simplifying reinforced feature selection via restructured choice strategy of single agent (ICDM20)
  • Self-optimizing feature generation via categorical hashing representation and hierarchical reinforcement crossing (ICDM23)
  • Continuous Optimization for Feature Selection with Permutation-Invariant Embedding and Policy-Guided Search (KDD25)
  • Fastft: Accelerating reinforced feature transformation via advanced exploration strategies (ICDE25)

D4AI-Time Series

We advance deep time series learning by tackling two fundamental challenges: regularities and distribution shift. We introduce deep expansion learning to capture periodic structures in time series for more accurate long-term forecasting, propose general paradigms to mitigate distribution shifts to ensure models remain reliable and adaptive across diverse and changing environments, and develop conditional generative flows to address time series imputation.

  • DEPTS: Deep Expansion Learning for Periodic Time Series Forecasting. (ICLR22)
  • Dish-TS: A General Paradigm for Alleviating Distribution Shift in Time Series Forecasting. (AAAI23)
  • Reconstructing missing variables for multivariate time series forecasting via conditional generative flows (IJCAI2024)
  • IN-Flow: Address Distribution Shift in Time Series Forecasting. (KDD2025)

D4AI-Geospatial

This project develops knowledge-guided geospatial representation learning by designing models that capture spatial substructures, temporal dependencies, and region-level correlations. It advances representation learning with innovations such as mutual information maximization, adversarial training, and reinforcement-driven imitation to improve profiling, prediction, and embedding in mobile and spatiotemporal networks. Together, these works form a unified approach that integrates peer interactions, temporal dynamics, and multi-view spatial structures for robust and interpretable geospatial intelligence.

  • Exploiting mutual information for substructure-aware graph representation learning (IJCAI20)
  • Adversarial Substructured Representation Learning for Mobile User Profiling (KDD19)
  • Peer and Temporal-Aware Representation Learning for Driving Behavior Analysis (KDD18)
  • Reinforced imitative graph representation learning for mobile user profiling: An adversarial training perspective (AAAI21)
  • Collective representation learning on spatiotemporal heterogeneous information networks (SIGSpatial19)
  • Spatiotemporal Representation Learning for Driving Behavior Analysis: A Joint Perspective of Peer and Temporal Dependencies (IEEE TKDE)
  • Unifying Inter-region Autocorrelation and Intra-region Structures for Spatial Embedding via Collective Adversarial Learning (KDD19)
  • Efficient Region Embedding with Multi-View Spatial Networks: A Perspective of Locality-Constrained Spatial Autocorrelations (AAAI19)

Conference Deadlines: ccfddl.github.io

  1. IJCAI (https://2025.ijcai.org/), SIGIR (https://sigir2025.dei.unipd.it/)
  2. KDD (https://kdd2024.kdd.org), ICML (https://icml.cc) UbiComp (https://www.ubicomp.org/ubicomp-iswc-2025/), ACL (https://2025.aclweb.org/)
  3. ICCV (https://internationalconferencealerts.com/eventdetails.php?id=2236516), ECML-PKDD (https://2024.ecmlpkdd.org), COLM(https://colmweb.org/), VLDB(https://www.vldb.org/2025/)
  4. ACM MM (https://acmmm2025.org/), SIGMOD (https://2026.sigmod.org/), VLDB(https://www.vldb.org/2025/),
  5. ACM CIKM Research/Applied  (https://cikm2024.org), VLDB(https://www.vldb.org/2025/), NeurIPS (https://neurips.cc), SIGSpatial (https://sigspatial2025.sigspatial.org/research-submission/), EMNLP (https://2025.emnlp.org/)
  6. IEEE ICDM(https://www3.cs.stonybrook.edu/~icdm2025/index.html), VLDB(https://www.vldb.org/2025/)
  7. VLDB(https://www.vldb.org/2025/)
  8. WSDM (https://www.wsdm-conference.org/2024/), ACM SIGKDD (https://kdd2024.kdd.org), BIBM (https://ieeebibm.org/BIBM2025/), VLDB(https://www.vldb.org/2025/), ICDE (https://ieee-icde.org/2025/cfp/)
  9. AAAI (https://aaai.org/aaai-conference/), ICLR (https://iclr.cc/Conferences/2026/Dates), VLDB(https://www.vldb.org/2025/)
  10. WWW (https://www2024.thewebconf.org), SIAM DM (https://www.siam.org/conferences/cm/conference/sdm24), VLDB(https://www.vldb.org/2025/),
  11. CVPR (https://cvpr.thecvf.com), ACL (https://2024.aclweb.org), VLDB(https://www.vldb.org/2025/), ICDE (https://ieee-icde.org/2025/cfp/)
  12. IEEE BigData (https://www3.cs.stonybrook.edu/~ieeebigdata2024/), NAACL (https://2024.naacl.org), VLDB(https://www.vldb.org/2025/)
Ira A. Fulton Schools of Engineering, Arizona State University
Maps and Locations Jobs Directory Contact ASU My ASU
Repeatedly ranked #1 on 30+ lists in the last 3 years
Copyright and Trademark Accessibility Privacy Terms of Use Emergency