Disentangled Representation Learning

Overview

Disentangled representation learning studies how to encode separate factors—such as content and style, identity and viewpoint, or disease and acquisition conditions—in distinct parts of a representation. Common approaches include variational autoencoders, contrastive objectives, sparse or orthogonal subspaces, causal models, and symmetry-based methods; many applications use the resulting factors for editing, robust prediction, or transfer across domains. Work now spans images, speech, graphs, time series, medicine, and robotics, with increasing attention to interventions and practical controllability. A central tension remains: factors are generally not identifiable from unlabelled data alone without assumptions or useful supervision, and standard disentanglement metrics can fail when their assumptions do not hold.

Latest papers

CardsList

245 papers10 papers in the last four weeks, down 63% on the four weeks before. 0.1% of all new papers.

  1. JEPA Dots: Dual-Latent World Model for Language-Guided Robot Control

    Nov 18, 2026Xianhao Pan, Chi Li, Hanchao Li +3Language-Conditioned Robot ControlWorld Models for VLAs

  2. Integrating Latent Block Models into Variational Autoencoders for Disentangled Representation Learning

    Nov 15, 2026Anasse Essalih, Reda Khoufache, Laurent Durville +5Latent Variable ModelsVariational Autoencoders

  3. Factorized Self-Supervised Speech Tokenization

    Nov 13, 2026Benjamin Van Niekerk, Jean-Philippe Letendre, Nicol Visser +3Audio TokenizationDiscrete Speech Representation Learning

  4. EASTAR: Event-Anchored Semantic and Task-Adaptive Region Representation Learning

    Nov 11, 2026Peng Liang, Genan Dai, Bowen ZhangUrban Representation LearningSemantic Representation Learning

  5. Exploring Post-Hoc Disentanglement of Audio Foundation Models for Music Similarity

    Nov 10, 2026Lucas de Souza Brandão, Gabriel da Costa Merlin, Diego Furtado SilvaRepresentation DisentanglementDisentangled Representation Learning

  6. OCCAMS: A Multi-Task Learning-Based Investigation of the CMB Power Spectra

    Oct 29, 2026Ankit Mandal, Rahul Shah, Md. Wali Hossain +3AstrophysicsDisentangled Representation Learning

  7. Neural-DISCO: Source-Conditioned Counterfactual Editing of Neural Population Activity

    Oct 23, 2026Nico Policzer, James Ho, Joseph Soo +1Neural EncodingCounterfactual Learning

  8. Decoupled Representation Learning for Face Presentation Attack Detection under Smart Glass Capture

    Oct 20, 2026Raghavendra Ramachandra, Narayan Vetrekar, Neha Naik +1Face Presentation Attack DetectionDisentangled Representation Learning

  9. Beyond Invariance: Factor-wise Selective Treatment of Shifts for Multimodal Learning

    Oct 15, 2026Guimin Hu, Qiwei Peng, Boyan Xu +1Distribution Shift RobustnessMultimodal Robustness

  10. CoDeFont: One-Shot Font Generation via Multi-Scale Factor-Specific Representation Learning

    Oct 15, 2026Rejoy Chakraborty, Prasun Roy, Saumik Bhattacharya +1Image GenerationDisentangled Representation Learning

  11. DEFT: Disentanglement-Enhanced Fine-Tuning for EEG Foundation Models

    Oct 10, 2026Yuqiu Deng, Muyun Jiang, Zongpeng Zhang +8EEG Foundation ModelsCross-Subject EEG Decoding

  12. Disentangling Task-Relevant Latent Dynamics for Reward Modeling

    Oct 9, 2026Dayou Li, Jionghao Wang, Peihao Wang +9Dense Reward RLMultimodal Reward Modeling

  13. Neural networks are more modular than single neurons suggest

    Oct 9, 2026Shreya Saha, Benjamin Bergen, Meenakshi KhoslaModular Neural NetworksNeural Network Interpretability

  14. CausalDreamer: Learning Predictive World Models with Latent Disentanglement

    Oct 8, 2026Prince Jha, Nils Lukas, Kun Zhang +1World Model LearningCausal World Models

  15. Disentangling Linguistic and Paralinguistic Information with Routed Sparse Autoencoders

    Oct 7, 2026Beimnet Bekele Guta, Xiaoyu Yang, Guangzhi Sun +1Disentangled Representation LearningSelf-Supervised Speech Representation Learning

  16. LinSlot: Exploiting Linear Representation hypothesis for unsupervised attribute discovery from slot based object representation

    Oct 7, 2026Sanket Gandhi, Utkarsh Giri, Varun Subramaniam +2Disentangled Representation LearningUnsupervised Learning

  17. DSReg: Provably Recovering Individual World Latents without Reconstruction

    Oct 7, 2026Yujia Zheng, David Klindt, Randall Balestriero +1Representation IdentifiabilityLatent World Models

  18. Learning Disentangled Representations with Quantum Variational Autoencoders

    Oct 5, 2026Gaoyuan Wang, Jerry Tan, Mark GersteinVariational AutoencodersDisentangled Representation Learning

  19. Operational Abstractions of Neural Network Concepts via Topological Representations

    Oct 4, 2026Mathieu Pont, Christoph GarthDisentangled Representation LearningModel Editing

  20. Towards Fast and Disentangled Counterfactuals for Visual Foundation Models

    Oct 1, 2026Sidney Bender, Benedikt Kunz, Ahmed Zeid +3Disentangled Representation LearningCounterfactual Explanations

  21. Values as Style: Disentangling Values from Semantics with One-Way Mixing for Low-Damage LLM Steering

    Sep 30, 2026Jiale Dai, Hongcan Deng, Liuxian Ma +2Disentangled Representation LearningLLM Alignment

  22. SAGE: Salient Factor Discovery and Generation with Visual Foundation Representations

    Sep 30, 2026Shuang Liang, Lejun Liao, Shiyuan Zhang +5Disentangled Representation LearningConditional Image Generation

  23. Invariant Atoms: Sparse Coordinates of Local Semantic Geometry in Language Model Representations

    Sep 29, 2026Muhammad Ahtesham, Xin ZhongRepresentation Geometry in Language ModelsInvariant Representation Learning