{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T12:44:49Z","timestamp":1782823489778,"version":"3.54.5"},"reference-count":74,"publisher":"American Institute of Mathematical Sciences (AIMS)","issue":"0","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MFC"],"published-print":{"date-parts":[[2026]]},"DOI":"10.3934\/mfc.2026024","type":"journal-article","created":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T11:53:34Z","timestamp":1782820414000},"page":"1-26","source":"Crossref","is-referenced-by-count":0,"title":["Diffusion implicit variational inference for complex posterior modeling"],"prefix":"10.3934","volume":"14","author":[{"given":"Lei","family":"Cai","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jian","family":"Xu","sequence":"additional","affiliation":[{"name":"RIKEN Center for Advanced Intelligence Project (AIP), RIKEN, Tokyo, 103-0027, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"2321","reference":[{"key":"key-10.3934\/mfc.2026024-1","unstructured":"<p>T. Akhound-Sadegh, J. Rector-Brooks, A. Bose, S. Mittal, P. Lemos, C.-H. Liu, M. Sendera, S. Ravanbakhsh, G. Gidel and Y. Bengio, <i>Iterated Denoising Energy Matching for Sampling From Boltzmann Densities<\/i>, arXiv, arXiv: 2402.06121.<\/p>"},{"key":"key-10.3934\/mfc.2026024-2"},{"key":"key-10.3934\/mfc.2026024-3","unstructured":"<p>J. Berner, L. Richter, M. Sendera, J. Rector-Brooks and N. Malkin, <i>From Discrete-Time Policies to Continuous-Time Diffusion Samplers: Asymptotic Equivalences and Faster Training<\/i>, arXiv, arXiv: 2501.06148.<\/p>"},{"key":"key-10.3934\/mfc.2026024-4","unstructured":"<p>C. Bishop and M. Tipping, <i>Variational Relevance Vector Machines<\/i>, arXiv, arXiv: 1301.3838.<\/p>"},{"key":"key-10.3934\/mfc.2026024-5","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2017.1285773"},{"key":"key-10.3934\/mfc.2026024-6","unstructured":"<p>Y. Burda, R. Grosse and R. Salakhutdinov, <i>Importance Weighted Autoencoders<\/i>, arXiv, arXiv: 1509.00519.<\/p>"},{"key":"key-10.3934\/mfc.2026024-7"},{"key":"key-10.3934\/mfc.2026024-8","doi-asserted-by":"publisher","DOI":"10.1109\/JAS.2019.1911645"},{"key":"key-10.3934\/mfc.2026024-9"},{"key":"key-10.3934\/mfc.2026024-10"},{"key":"key-10.3934\/mfc.2026024-11","unstructured":"<p>R. Child, <i>Very Deep Vaes Generalize Autoregressive Models and can Outperform them on Images<\/i>, arXiv, arXiv: 2011.10650.<\/p>"},{"key":"key-10.3934\/mfc.2026024-12","unstructured":"<p>R. Child, S. Gray, A. Radford and I. Sutskever, <i>Generating Long Sequences with Sparse Transformers<\/i>, arXiv, arXiv: 1904.10509.<\/p>"},{"key":"key-10.3934\/mfc.2026024-13","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3261988"},{"key":"key-10.3934\/mfc.2026024-14"},{"key":"key-10.3934\/mfc.2026024-15"},{"key":"key-10.3934\/mfc.2026024-16","unstructured":"<p>L. Dinh, D. Krueger and Y. Bengio, <i>Nice: Non-linear Independent Components Estimation<\/i>, arXiv, arXiv: 1410.8516.<\/p>"},{"key":"key-10.3934\/mfc.2026024-17","doi-asserted-by":"crossref","unstructured":"<p>H. Feng, S. Hou, L.-Y. Wei and D.-X. Zhou, Cnn models for readability of chinese texts, <i>Mathematical Foundations of Computing<\/i>, <b>5<\/b>.<\/p>","DOI":"10.3934\/mfc.2022021"},{"key":"key-10.3934\/mfc.2026024-18"},{"key":"key-10.3934\/mfc.2026024-19","doi-asserted-by":"publisher","DOI":"10.1109\/JAS.2021.1004108"},{"key":"key-10.3934\/mfc.2026024-20","unstructured":"<p>L. Hazami, R. Mama and R. Thurairatnam, <i>Efficientvdvae: Less is more<\/i>, arXiv, arXiv: 2203.13751.<\/p>"},{"key":"key-10.3934\/mfc.2026024-21"},{"key":"key-10.3934\/mfc.2026024-22","unstructured":"<p>M. Hoffman, D. Blei, C. Wang and J. Paisley, Stochastic variational inference, <i>Journal of Machine Learning Research<\/i>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-23","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(89)90020-8"},{"key":"key-10.3934\/mfc.2026024-24","unstructured":"<p>X. Huang, H. Dong, Y. Hao, Y.-A. Ma and T. Zhang, <i>Reverse Diffusion Monte Carlo<\/i>, arXiv, arXiv: 2307.02037.<\/p>"},{"key":"key-10.3934\/mfc.2026024-25","unstructured":"<p>F. Husz\u00e1r, <i>Variational Inference using Implicit Distributions<\/i>, arXiv, arXiv: 1702.08235.<\/p>"},{"key":"key-10.3934\/mfc.2026024-26"},{"key":"key-10.3934\/mfc.2026024-27","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007665907178"},{"key":"key-10.3934\/mfc.2026024-28"},{"key":"key-10.3934\/mfc.2026024-29","unstructured":"<p>D. Kingma, <i>Adam: A Method for Stochastic Optimization<\/i>, arXiv, arXiv: 1412.6980.<\/p>"},{"key":"key-10.3934\/mfc.2026024-30","unstructured":"<p>D. Kingma, T. Salimans, R. Jozefowicz, X. Chen, I. Sutskever and M. Welling, Improved variational inference with inverse autoregressive flow, <i>Advances in Neural Information Processing Systems<\/i>, <b>29<\/b>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-31","unstructured":"<p>D. Kingma and M. Welling, <i>Auto-Encoding Variational Bayes<\/i>, arXiv, arXiv: 1312.6114.<\/p>"},{"key":"key-10.3934\/mfc.2026024-32"},{"key":"key-10.3934\/mfc.2026024-33","unstructured":"<p>J. Lim and A. Johansen, <i>Particle Semi-Implicit Variational Inference<\/i>, arXiv, arXiv: 2407.00649.<\/p>"},{"key":"key-10.3934\/mfc.2026024-34","unstructured":"<p>X. Ma, X. Kong, S. Zhang and E. Hovy, Macow: Masked convolutional generative flow, <i>Advances in Neural Information Processing Systems<\/i>, <b>32<\/b>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-35","unstructured":"<p>X. Ma, C. Zhou and E. Hovy, <i>Mae: Mutual Posterior-Divergence Regularization for Variational Autoencoders<\/i>, arXiv, arXiv: 1901.01498.<\/p>"},{"key":"key-10.3934\/mfc.2026024-36","unstructured":"<p>L. Maal\u00f8e, M. Fraccaro, V. Li\u00e9vin and O. Winther, Biva: A very deep hierarchy of latent variables for generative modeling, <i>Advances in Neural Information Processing Systems<\/i>, <b>32<\/b>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-37","unstructured":"<p>J. Menick and N. Kalchbrenner, <i>Generating high Fidelity Images with Subscale Pixel Networks and Multidimensional Upscaling<\/i>, arXiv, arXiv: 1812.01608.<\/p>"},{"key":"key-10.3934\/mfc.2026024-38"},{"key":"key-10.3934\/mfc.2026024-39","unstructured":"<p>V. Moens, H. Ren, A. Maraval, R. Tutunov, J. Wang and H. Ammar, <i>Efficient Semi-Implicit Variational Inference<\/i>, arXiv, arXiv: 2101.06070.<\/p>"},{"key":"key-10.3934\/mfc.2026024-40"},{"key":"key-10.3934\/mfc.2026024-41","unstructured":"<p>G. Papamakarios, T. Pavlakou and I. Murray, Masked autoregressive flow for density estimation, <i>Advances in Neural Information Processing Systems<\/i>, <b>30<\/b>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-42","unstructured":"<p>A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein and L. Antiga, Pytorch: An imperative style, high-performance deep learning library, <i>Advances in Neural Information Processing Systems<\/i>, <b>32<\/b>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-43","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i19.34194"},{"key":"key-10.3934\/mfc.2026024-44","unstructured":"<p>A. Razavi, A. Oord, B. Poole and O. Vinyals, <i>Preventing Posterior Collapse with Delta-Vaes<\/i>, arXiv, arXiv: 1901.03416.<\/p>"},{"key":"key-10.3934\/mfc.2026024-45","doi-asserted-by":"publisher","DOI":"10.1109\/JAS.2024.124290"},{"key":"key-10.3934\/mfc.2026024-46"},{"key":"key-10.3934\/mfc.2026024-47","unstructured":"<p>H. Sadeghi, E. Andriyash, W. Vinci, L. Buffoni and M. Amin, <i>Pixelvae++: Improved Pixelvae with Discrete Prior<\/i>, arXiv, arXiv: 1908.09948.<\/p>"},{"key":"key-10.3934\/mfc.2026024-48","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-statistics-010814-020120"},{"key":"key-10.3934\/mfc.2026024-49"},{"key":"key-10.3934\/mfc.2026024-50"},{"key":"key-10.3934\/mfc.2026024-51","unstructured":"<p>J. Shi, S. Sun and J. Zhu, <i>Kernel Implicit Variational Inference<\/i>, arXiv, arXiv: 1705.10119.<\/p>"},{"key":"key-10.3934\/mfc.2026024-52"},{"key":"key-10.3934\/mfc.2026024-53","unstructured":"<p>C. S\u00f8nderby, T. Raiko, L. Maal\u00f8e, S. S\u00f8nderby and O. Winther, Ladder variational autoencoders, <i>Advances in Neural Information Processing Systems<\/i>, <b>29<\/b>.<\/p>"},{"key":"key-10.3934\/mfc.2026024-54","unstructured":"<p>Y. Song and D. Kingma, <i>How to Train your Energy-Based Models<\/i>, arXiv, arXiv: 2101.03288.<\/p>"},{"key":"key-10.3934\/mfc.2026024-55","unstructured":"<p>Y. Tian, N. Panda and Y. Lin, <i>Liouville Flow Importance Sampler<\/i>, arXiv, arXiv: 2405.06672.<\/p>"},{"key":"key-10.3934\/mfc.2026024-56"},{"key":"key-10.3934\/mfc.2026024-57"},{"key":"key-10.3934\/mfc.2026024-58"},{"key":"key-10.3934\/mfc.2026024-59"},{"key":"key-10.3934\/mfc.2026024-60","unstructured":"<p>F. Vargas, W. Grathwohl and A. Doucet, <i>Denoising Diffusion Samplers<\/i>, arXiv, arXiv: 2302.13834.<\/p>"},{"key":"key-10.3934\/mfc.2026024-61"},{"key":"key-10.3934\/mfc.2026024-62","doi-asserted-by":"crossref","unstructured":"<p>M. Wainwright, M. Jordan, et al., Graphical models, exponential families, and variational inference, <i>Foundations and Trends\u00ae in Machine Learning<\/i>, <b>1<\/b> (2008), 1-305.<\/p>","DOI":"10.1561\/2200000001"},{"key":"key-10.3934\/mfc.2026024-63"},{"key":"key-10.3934\/mfc.2026024-64","doi-asserted-by":"crossref","unstructured":"<p>Y. Xia, J. Zhou, T. Xu and W. Gao, An improved deep convolutional neural network model with kernel loss function in image classification, <i>Mathematical Foundations of Computing<\/i>, <b>3<\/b>.<\/p>","DOI":"10.3934\/mfc.2020005"},{"key":"key-10.3934\/mfc.2026024-65"},{"key":"key-10.3934\/mfc.2026024-66","unstructured":"<p>R. Yao and Y. Yang, <i>Mean-Field Variational Inference Via Wasserstein Gradient Flow<\/i>, arXiv, arXiv: 2207.08074.<\/p>"},{"key":"key-10.3934\/mfc.2026024-67"},{"key":"key-10.3934\/mfc.2026024-68"},{"key":"key-10.3934\/mfc.2026024-69","unstructured":"<p>L. Yu and C. Zhang, <i>Semi-Implicit Variational Inference Via Score Matching<\/i>, arXiv, arXiv: 2308.10014.<\/p>"},{"key":"key-10.3934\/mfc.2026024-70","doi-asserted-by":"publisher","DOI":"10.3934\/mfc.2025021"},{"key":"key-10.3934\/mfc.2026024-71","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2889774"},{"key":"key-10.3934\/mfc.2026024-72","unstructured":"<p>Q. Zhang and Y. Chen, <i>Path Integral Sampler: A Stochastic Control Approach for Sampling<\/i>, arXiv, arXiv: 2111.15141.<\/p>"},{"key":"key-10.3934\/mfc.2026024-73"},{"key":"key-10.3934\/mfc.2026024-74","unstructured":"<p>D. Zou, P. Xu and Q. Gu, Stochastic gradient hamiltonian monte carlo methods with recursive variance reduction, <i>Advances in Neural Information Processing Systems<\/i>, <b>32<\/b>.<\/p>"}],"container-title":["Mathematical Foundations of Computing"],"original-title":[],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T11:54:42Z","timestamp":1782820482000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.aimsciences.org\/\/article\/doi\/10.3934\/mfc.2026024"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":74,"journal-issue":{"issue":"0","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.3934\/mfc.2026024","relation":{},"ISSN":["2577-8838"],"issn-type":[{"value":"2577-8838","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}