Deep generative models have been enjoying success in modeling continuous data. However it remains challenging to capture the representations for discrete structures with formal grammars and semantics, e.g., computer programs and molecular structures. How to generate both syntactically and semantically correct data still remains largely an open problem. Inspired by the theory of compiler where the syntax and semantics check is done via syntax-directed translation (SDT), we propose a novel syntax-directed variational autoencoder (SD-VAE) by introducing stochastic lazy attributes. This approach converts the offline SDT check into on-the-fly generated guidance for constraining the decoder. Comparing to the state-of-the-art methods, our approach enforces constraints on the output space so that the output will be not only syntactically valid, but also semantically reasonable. We evaluate the proposed model with applications in programming language and molecules, including reconstruction and program/molecule optimization. The results demonstrate the effectiveness in incorporating syntactic and semantic constraints in discrete generative models, which is significantly better than current state-of-the-art approaches.
CAVACHON: a hierarchical variational autoencoder to integrate multi-modal single-cell data
Centre for Molecular Medicine Norway (NCMM)(挪威分子医学中心(NCMM));Nordic EMBL Partnership(北欧欧洲分子生物学实验室合作组织);University of Oslo(奥斯陆大学);National Taiwan University(国立台湾大学);Oslo University Hospital(奥斯陆大学医院);CISPA Helmholtz Center for Information Security(CISPA亥姆霍兹信息安全中心)
University of British Columbia(不列颠哥伦比亚大学);Electronic Arts Vancouver(艺电温哥华分部)
A fundamental problem in computer animation is that of realizing purposeful and realistic human movement given a sufficiently-rich set of motion capture clips. We learn data-driven generative models of human movement using autoregressive conditional variational autoencoders, or Motion VAEs. The latent variables of the learned autoencoder define the action space for the movement and thereby govern its evolution over time. Planning or control algorithms can then use this action space to generate desired motions. In particular, we use deep reinforcement learning to learn controllers that achieve …
Heterogeneous Directed Hypergraph Neural Network over abstract syntax tree (AST) for Code Classification
School of Computer Science and Technology, East China Normal University(计算机科学与技术学院,东华大学);NPPA Key Laboratory of Publishing Integration Development, ECNUP(出版集成发展关键实验室,ECNUP)
ControlTraj: Controllable Trajectory Generation with Topology-Constrained Diffusion Model
Southern University of Science and Technology(南方科技大学);City University of Hong Kong(香港城市大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));York University(约克大学)
Center for Computing Research, Sandia National Laboratories(桑迪亚国家实验室计算研究中心);Applied Information Sciences Center, Sandia National Laboratories(桑迪亚国家实验室应用信息科学中心);School of Computing and Augmented Intelligence, Arizona State University(亚利桑那州立大学计算与增强智能学院);School of Engineering and Applied Science, University of Pennsylvania(宾夕法尼亚大学工程与应用科学学院)
Directed Metric Structures arising in Large Language Models
INRIA(法国国家信息与自动化研究所);CMAP, École polytechnique, IP Paris(巴黎综合理工学院CMAP实验室(IP巴黎));CNRS(法国国家科学研究中心);ATHENA Research Center(雅典娜研究中心);Institute for Language and Speech Processing(语言与语音处理研究所);IHES(高等科学研究所)