Latent Space Alignment

Ask about this topic

Overview

Latent space alignment studies how to make representations from different models, modalities, domains, or people comparable and useful together. Common approaches learn linear or orthogonal maps with Procrustes methods, use optimal transport or shared semantic anchors, or regularize representations to preserve geometry and task-relevant structure. These methods support tasks such as model stitching, cross-modal retrieval, domain adaptation, and cross-subject brain decoding, evaluated through measures including retrieval accuracy, activation reconstruction, and downstream task performance. A central tension is that geometric or distributional similarity does not guarantee functional compatibility: alignment can preserve nuisance information, depend on paired examples, or improve a similarity metric without improving the task.

Latest papers

CardsList

76 papers20 papers in the last four weeks, up 122% on the four weeks before. 0.1% of all new papers.

  1. Five Architectures, One Shared Space Cross-Model Representational Alignment Without Weight Merging

    Oct 24, 2026Oladimeji Anthonio, Dimeji Olawuyi, Oloruntoba Ajayi +1Latent Space AlignmentRepresentation Geometry in Language Models

    Centre for Clinical Intelligence and Safety, `Iy`aw´o · `Iy`aw´o · Faculty of Clinical Sciences, College of Medicine, University of Ibadan

  2. The Geometry of Meaning in Neural Manifolds

    Oct 22, 2026Valentino MaiorcaNeural Representation GeometrySemantic Correspondence

    Department of Computer Science, Faculty of Information Engineering, Informatics, and Statistics, Sapienza University of Rome

  3. Towards Consistent Geospatial Representations: Aligning One Encoder to Many Foundation Models

    Oct 20, 2026Arthur Ouaknine, David RolnickGeospatial Foundation ModelsGeospatial Representation Learning

    McGill University · Mila – Quebec AI Institute

  4. MiCLA: Mixture-to-Constituent Latent Alignment for Polyphonic Audio Representation Learning

    Oct 20, 2026Kohei Yamamoto, Kosuke OkusaAudio Representation LearningSelf-Supervised Pre-Training

    Research & Development Center, Technology Division, Oki Electric Industry Co., Ltd., Saitama, Japan. · Department of Data Science for Business Innovation, Chuo University, Tokyo, Japan.

  5. CASE: Complementary Articulation-Aware Speech Encoder to Enhance Speech-Driven 3D Facial Animation

    Oct 18, 2026Jaehong Myung, Seungho Eum, Taegon Kim +2Audio-Driven Facial AnimationAudio-Visual Representation Learning

    Department of Artificial Intelligence, Sogang University, Seoul, Republic of Korea · Department of Computer Science and Engineering, Sogang University, Seoul, Republic of Korea

  6. Native Geometry vs. Linear Recoverability: How Alignment Changes Platonic Scaling

    Oct 15, 2026Angus Hoy, Austin Lutterbach, Jasper Chong +2Neural Representation GeometryLatent Space Geometry

    EleutherAI SOAR · UniverseTBD · University of Queensland +4

  7. Semantic Latent Alignment for Flow Matching with Invertible Bi-Lipschitz Adapters

    Oct 15, 2026Felix Nützel, Mischa Dombrowski, Luca Hagen +1Latent Space AlignmentConditional Image Generation

    Friedrich-Alexander-Universität Erlangen-Nürnberg · Imperial College London

  8. From Plato to Aristotle: Toward a General Theory of Representation Hypotheses

    Oct 14, 2026Dai Shi, Andi Han, Xiaoyu Li +2Representation AlignmentPlatonic Representation Hypothesis

    University of Cambridge · University of Sydney · University of New South Wales

  9. Train to Stitch: Topology-Preserving Regularization for Interoperable Representations

    Oct 14, 2026Akshay Malhotra, Nicolas Zilberstein, Shahab Hamidi-RadLatent Space AlignmentAutoencoders

    InterDigital Communications Inc., Los Altos, CA, USA · Rice University, Houston, TX, USA

  10. What Makes a Latent Space Shared? Direct and Recoverable Compatibility in Motion VAEs

    Oct 12, 2026Iago Brito, Walcy Rios, Lucas Fernandes +2Latent Space AlignmentMotion Representation Learning

    Advanced Knowledge Center in Immersive Technologies (AKCIT) Federal University of Goiás (UFG)

  11. MSVAE: Rethinking Musical Semantics in Continuous Latent Spaces

    Oct 12, 2026Guobin Ma, Yuepeng Jiang, Dake Guo +15Audio Representation LearningCompositional Representation Learning

    Geely Automobile Research Institute (Ningbo) Company Ltd, China · Independent Researcher, China · Faculty of Electronic and Information Engineering, Xi’an Jiaotong University, China +1

  12. Align-VAE: Aligning Attention and Geometry in VecSet VAEs for 3D Reconstruction

    Oct 10, 2026Huaizhi Qu, Mingxin Yang, Xintong Han +2Variational Autoencoders3D Reconstruction

    The Hong Kong University of Science and Technology · Tencent Hunyuan

  13. ROSETTA: Rotation Subspace Estimation for Training-free Transmodal Alignment

    Oct 9, 2026Alex Junho LeeCross-Modal AlignmentLatent Space Alignment

    School of Mechanical Systems Engineering, Sookmyung Women’s University, Seoul 04310, South Korea

  14. Local Prototype Reconstruction for Text-Compatible Speech-to-LLM Bridge Pretraining

    Oct 8, 2026Xinnian Zhao, Chia-Hua Wu, Pu Wang +1Speech Language ModelsLatent Space Alignment

    Department of Electrical Engineering (ESAT), KU Leuven, Leuven, Belgium · Institute of Information Science, Academia Sinica, Taiwan

  15. Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss

    Oct 7, 2026Ruoyu Zhao, Yuting Chen, Jinheng Zhang +2Contrastive LearningMulti-View Contrastive Learning

    City University of Hong Kong · Georgia Institute of Technology · University of Pennsylvania +2

  16. A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification

    Oct 7, 2026Yucheng Gong, Rui Zhou, Binbin Zeng +2Cross-Domain GeneralizationAudio Classification

    Department of Software Engineering, School of Software, Chongqing Finance and Economics College Chongqing Key Laboratory of Data Mining and Integrated Application for Ecological Environment

  17. Learning Cross-Model Activation Alignments with Explicit Many-to-Many Layer Maps

    Oct 6, 2026Alina Sudakov, Guy Bar-Shalom, Fabrizio Frasca +1Latent Space AlignmentTransfer Learning

    Technion · Technion, NVIDIA Research

  18. On Color Alignment in VAE Latent Spaces and Its Applications

    Oct 5, 2026Julian D. Santamaria, Kai Wang, Jesús Malo +2Variational AutoencodersImage Generation

    Computer Vision Center · Universitat Aut`onoma de Barcelona · City University of Hong Kong (Dongguan) +1

  19. Do Neural Networks Learn Structure-Preserving Maps? A Case Study in Latent-to-Hilbert Embeddings

    Oct 4, 2026Muhammad Adnan ShahzadRepresentation LearningLatent Space Alignment

    Department of Computer Science and Software Engineering (CSSE) Montreal, QC, Canada

  20. NinaXander: Feasibility and Limits of Composing Frozen Language Models Across Architecture Families via a Shared Latent Space

    Sep 29, 2026Takanori Kotama, Shun-ichiro Hayashi, Daichi Mukunoki +2Latent Space Alignment

    Graduate School of Informatics Nagoya University · Information Technology Center Nagoya University

  21. What Should Federated LoRA Share? FedSAIL via Input-aware Subspace Alignment

    Sep 26, 2026Junye Du, Shuaida He, Long FengFederated LearningLow-Rank Adaptation

    School of Computing & Data Science The University of Hong Kong

  22. WALT: Learning World-Model-Aligned Latent Trajectories for Autonomous Driving

    Sep 24, 2026Mingkai Jia, Jiaxin Guo, Zhijian Shu +5World Model-Based PlanningAutonomous Driving Planning

    The Hong Kong University of Science and Technology. · Horizon Robotics. · The Chinese University of Hong Kong. +2

  23. It's the Geometry, Not the Model: Effective Rank and Subspace Alignment in Functional Connectivity Classification

    Sep 24, 2026Xiao Fan, Jingyuan Li, Yubo Han +6Intrinsic DimensionalityFunctional Connectivity

  24. Stable Neural Decoding Across Sessions via Task-Conditioned Latent Alignment for Brain-Machine Interfaces

    Sep 23, 2026Canyang Zhao, Bolin Peng, J. Patrick Mayo +2Brain-Computer InterfacesNeural Decoding

    Institute of Automation, Chinese Academy of Sciences, Beijing, China · Departments of Ophthalmology and Bioengineering, University of Pittsburgh, Pittsburgh, USA · School of Science and Engineering, Chinese University of Hong Kong, Shenzhen, China

  25. LatentReRig: An SDF-Based VAE with Dual Decoders for Latent-Space Deformation Conditioning

    Sep 21, 2026Daniele Dolci, Fabrizio Poggioni, Carlo MelchiorriHuman Avatar AnimationLatent Space Alignment

    AnimSchool, Utah, USA · Mostly Very True SRL, Viterbo, Italy · Università degli Studi Guglielmo Marconi, Rome, Italy

  26. SPACE: Semantic Projection and Alignment of CLIP Embeddings for Domain Adaptation

    Sep 19, 2026João Renato Ribeiro Manesco, Danilo Samuel Jodas, Douglas Rodrigues +2Domain AdaptationUnsupervised Domain Adaptation

    São Paulo State University (UNESP) School of Sciences, Bauru-SP, Brazil

  27. Time series generation with spectrally aligned latent flow matching

    Sep 18, 2026Camilo Carvajal Reyes, Felipe TobarTime Series GenerationFlow-Based Generative Modeling

    Department of Mathematics, Imperial College London, 180 Queen’s Gate, London SW7 2HR, United Kingdom

  28. Multimodal Floorplan Encoding: Learning Dense Modality-Invariant Representations

    Sep 11, 2026Xavier Anadón, Rémi Pautrat, Rui WangLatent Space AlignmentCross-Modal Representation Learning

    University of Zaragoza, Spain · Microsoft, Switzerland

  29. Temporal Forcing: 4D Representation Alignment for Vision-Language-Action Models

    Aug 31, 2026Xingyu Ding, Yuzhong Zhao, Chunhai Zhao +3Representation LearningTemporal Representation Learning

    Nanjing University, Nanjing, China · Institute of Automation, Chinese Academy of Sciences, Beijing, China · University of Chinese Academy of Sciences, Beijing, China

  30. Unsupervised Latent Space Alignment with Hyperspherical Geodesic Matching

    Aug 28, 2026Cameron Ryan, Vivek Sivaraman Narayanaswamy, Kowshik Thopalli +1Latent Space AlignmentLatent Space Geometry

    Northeastern University, Boston, MA 02115. Work performed during a summer internship at Lawrence Livermore National Laboratory · Lawrence Livermore National Laboratory, Livermore, CA 94551-0808.