Autoregressive Generation

Ask about this topic

Overview

Autoregressive generation models how to build sequences one token, event, or chunk at a time, using prior outputs to create text, audio, images, motion, graphs, molecules, and other structured data. Common approaches include next-token transformers, hierarchical or coarse-to-fine prediction, compact learned tokenizations, and hybrids that combine autoregression with diffusion, flow matching, or reward-based optimization. Researchers use conditioning, structured representations, and training on imperfect rollouts to improve controllability, validity, and long-horizon consistency across benchmarks ranging from language and music to 3D geometry and scientific simulation. A central tension is that sequential dependencies support coherent, flexible outputs but can make generation slow and errors accumulate, motivating parallel decoding and better measures of quality beyond next-step prediction.

Latest papers

CardsList

109 papers16 papers in the last four weeks, up 60% on the four weeks before. 0.1% of all new papers.

  1. CrystalAR: Autoregressive Modeling for Symmetric Crystal Generation

    Oct 20, 2026Ziqiao Meng, Rongsheng Chen, Zhi Zheng +4Crystal Structure GenerationCrystal Structure Prediction

    Institute for Functional Intelligent Materials (IFIM), National University of Singapore · School of Computing, National University of Singapore · Southern University of Science and Technology

  2. OrderGPT: Generation Order as a Learnable Modeling Choice

    Oct 19, 2026Chenhe Yang, Yuhuan Lyu, Tianyang HuAutoregressive ModelingAutoregressive Generation

    SAGA Lab, The Chinese University of Hong Kong, Shenzhen

  3. FracTS: Hierarchical and Autoregressive Time Series Generation

    Oct 18, 2026Jiayu Li, Umair Afzal, Zilong Zhao +3Time Series GenerationAutoregressive Generation

    University of Illinois Urbana-Champaign · Betterdata AI · National University of Singapore

  4. TopoCompile: Molecular Generation with Exact Global Count Constraints

    Oct 17, 2026Xiang Wu, Rong-Hua Li, Quanqing Xu +2Molecular GenerationMolecular Graph Generation

  5. Power-Law Correlations Without Apparent Temperature-Tuned Criticality in Fixed-Tree Hierarchical Languages

    Oct 17, 2026Marcel Kühn, Bernd RosenowAutoregressive Generation

    Institute for Theoretical Physics, Leipzig University · Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI), Leipzig University

  6. Thought Blocks: Non-Autoregressive Latent Communication for Multi-Agent Systems via Trajectory-to-State Compilation

    Oct 16, 2026Zhenhan Guan, Xiaolong Liu, Hengwei Ye +1Multi-Agent CommunicationMulti-Agent LLM Systems

    ShanghaiTech University, Shanghai, China · AI Lab, The Yangtze River Delta, Changzhou, China

  7. MARS - A foundational map auto-regressor

    Oct 14, 2026Qi Zhang, Suvam Bag, Rupanjali Kukal +9Geospatial Foundation ModelsHD Map Construction

    Microsoft

  8. Limitations on Autoregressive Token Generation via Sequential Sampling

    Oct 10, 2026Gal Arnon, Andrej Bogdanov, Paul Lou +3Autoregressive Generation

    NTT Research · University of Ottawa · Bocconi University

  9. Sonic-AMG: Streaming Human Motion Generation from Spatial Audio

    Oct 9, 2026Jingcheng Ni, Zekun Li, Zhiyang Dou +6Human Motion GenerationAutoregressive Generation

    Brown University · MIT · Yale University +1

  10. The Lattice of Transition Laws

    Oct 8, 2026T. Y. Tsui, Jiatao Gu, Lingjie LiuParallel DecodingDiffusion Sampling

    University of Pennsylvania

  11. Empirical Variational Autoencoder

    Oct 5, 2026Kaede ShioharaVariational AutoencodersLatent Variable Models

    The University of Tokyo

  12. Scaling Collider Event Generation with Residual-Quantized Tokens

    Sep 30, 2026Dan Godi, Dmitrii Kobylianskii, Eilam GrossAutoregressive GenerationConditional Generative Modeling

    Weizmann Institute of Science, Rehovot, Israel

  13. Autoregressive Frontier Expansion: Growing Trees with Graph Machine Learning

    Sep 29, 2026Umer Gupta, Saku Peltonen, Martin RitzertGraph GenerationAutoregressive Generation

    Independent Researcher London, United Kingdom · ETH Zurich Zurich, Switzerland · Leipzig University Leipzig, Germany

  14. GraphVQ: Structure-Aware Autoregressive Decoding over Context-Quantized Graph Tokens

    Sep 29, 2026Yuxiang Yao, Zijun ZhaoAutoregressive GenerationGraph Foundation Models

    Research Center, China Life Insurance Company Ltd. · School of Computer Science and Technology Beijing Institute of Technology

  15. DrawingsDreamer: A Unified Multi-View Engineering Drawings Generation Model

    Sep 28, 2026Shurui Liu, Weide Chen, Changwang Yi +1Vector Graphics GenerationGenerative Modeling

    School of Computer Science and Engineering, Sun Yat-sen University, China · School of Intelligent Systems Engineering, Shenzhen Campus of Sun Yat-sen University, China

  16. HALO: Enhancing Time Series Generation via Hyperspherical Latents and Masked AutoregRessive Modeling

    Sep 28, 2026Chunyi Hou, Xiangfei Qiu, Hanyin Cheng +2Variational AutoencodersTime Series Generation

    School of Data Science and Engineering, East China Normal University, Shanghai, China

  17. Shared Autoregressive Context Can Distort Relationships in Synthetic Data

    Sep 26, 2026Thomas S. RobinsonSynthetic Data Generation for Language ModelsSynthetic Data Evaluation

    London School of Economics and Political Science

  18. Editable Map-Conditioned Trajectory Generation for Human Mobility Simulation

    Sep 26, 2026Takayuki Mizuno, Shouji Fujimoto, Mikito Hiruki +1Trajectory GenerationHuman Behavior Simulation

    National Institute of Informatics Tokyo, Japan · Kanazawa Gakuin University Kanazawa, Ishikawa, Japan · The University of Tokyo Tokyo, Japan

  19. ShapeLex: Decoupling Local Shape Symbolization and Global Scale Modeling for Text-Controlled Time Series Generation

    Sep 21, 2026Subo Wei, Jianqi Gao, Mingyan Fan +3Time Series GenerationAutoregressive Generation

    School of Computer Engineering and Science, Shanghai University, Shanghai, China · Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai, China

  20. SewFusion: Tailored Generation of Topology and Panel-Level Geometry for Sewing Patterns

    Sep 20, 2026Jiaxin Lin, Xiao Pan, Hangjie Yuan +3Generative ModelingAutoregressive Generation

    Digital Creative Technology Laboratory, Shenzhen University · Zhejiang University

  21. Divergence Timing and Cumulative Disagreement under KV-Cache Eviction

    Sep 15, 2026Xinyue Luo, Fei YuKV CachingKV-Cache Eviction

    Wealth AI Lab, Ant Group

  22. Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation

    Sep 11, 2026Thao Nguyen, Jeonghwan Kim, Zhenhailong Wang +1Molecular GenerationMolecule Design

    Siebel School of Computing and Data Science University of Illinois Urbana-Champaign Urbana, IL 61801, USA

  23. PAC-Private Autoregressive Generation: Calibrating Noise to Ensemble Disagreement

    Sep 4, 2026Mina Mirzadehsarcheshmeh, Amir Keyvan KhandaniPrivacy-Preserving Language ModelsPrivacy Leakage in Language Models

  24. Sparse auto-regressive modeling for scene generation from multi-view images

    Sep 3, 2026Thomas Lucas, Maxime Pietrantoni, Philippe Weinzaepfel +4Sparse-View 3D Reconstruction3D Scene Generation

    1NAVER LABS Europe, France · 2NAVER LABS, South Korea · 3Carnegie Mellon University, USA

  25. Geometry-aware Latent Autoregressive Generative Model for PDEs in Complex Domains

    Aug 31, 2026Zi Wang, Minghui Xu, Tapan MukerjiGraph Neural NetworksNeural PDE Solvers

    Department of Energy Science and Engineering Stanford University Stanford, CA 94305

  26. Preference Shapes Relevance: Cross-component Hierarchical Semantic Alignment for Personalized Generative Retrieval

    Aug 31, 2026Gaoming Zhang, Angqing Jiang, Jianchun Song +4Generative RetrievalSemantic Alignment

    University of Science and Technology of China · Meituan

  27. Generation of High-Level Concepts in 3D Scene Graphs via Autoregressive Diffusion

    Aug 28, 2026Jose Andres Millan-Romera, Samuel Cognolato, Holger Voos +23D SGGAutoregressive Generation

    Automation and Robotics Research Group, SnT University of Luxembourg, Luxembourg · University of Padova, Padova, Italy · Fondazione Bruno Kessler, Trento, Italy +1

  28. Reward-Guided Autoregressive Graph Generation for Efficient Multi-Agent Communication Topology Design

    Aug 20, 2026Poomphob Suwannapichat, Boonyarit Changaival, Caesar Wu +1Multi-Agent Communication TopologyGraph Generation

    University of Luxembourg, Luxembourg · King Mongkut’s University of Technology Thonburi, Thailand

  29. MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling

    Aug 14, 2026Manwen Liao, Xinyu Lian, Jian Mao +11Autoregressive 3D Scene GenerationPart-Level 3D Generation

    The University of Hong Kong · Shanghai Artificial Intelligence Laboratory · Fudan University +5

  30. MiDashengLM-Gen: Unified Audio Scene Generation via LLM-Driven Autoregressive Flow Matching

    Aug 12, 2026Xingwei Sun, Heinrich Dinkel, Gang Li +7Text-to-Audio GenerationAutoregressive Generation

    MiLM Plus, Xiaomi Inc., Beijing, China · X-LANCE Lab, Shanghai Jiao Tong University, Shanghai, China