Representation Disentanglement

Ask about this topic

Overview

Representation disentanglement aims to make different properties—such as content and style, identity and expression, or task-relevant and domain-specific signals—independently accessible in a model’s features. Common approaches include variational objectives, orthogonality and independence constraints, adversarial or contrastive learning, sparse autoencoders, and architectures with separate feature streams or subspaces. The methods are used in settings such as controllable image and speech generation, domain generalization, privacy and fairness, and mechanistic interpretability, with evaluation through probes, leakage tests, reconstruction quality, and downstream performance. A central challenge is that factors may be correlated or not uniquely identifiable, while suppressing unwanted information can also reduce useful task information or reconstruction fidelity.

Latest papers

CardsList

98 papers9 papers in the last four weeks, up 29% on the four weeks before. 0.1% of all new papers.

  1. Rethinking SAE Evaluation: An Atomic Interpretable Unit Framework Reveals Hidden Polysemanticity and Redundancy

    Oct 21, 2026Riteng Zhang, Xiaoqian WangSparse AutoencodersExplainability Evaluation

    Purdue University

  2. Isolating a Language Model's Judgment with a Label-Free Subspace

    Oct 15, 2026Chaemin Jang, Donggyu Lee, Jeesoo Kim +2LLM InterpretabilityRepresentation Disentanglement

    Korea Advanced Institute of Science and Technology (KAIST)

  3. DeTac: Dual-branch Decoupling of Texture and Force Signals in Vision-Based Tactile Sensors

    Oct 12, 2026Zhirui Zhang, Yunkai Tang, Jiahang Li +6Vision-Based Tactile SensingRepresentation Disentanglement

    State Key Laboratory of Multimedia Information Processing, National Engineering Research Center of Visual Technology, and PKU-AI2 Robotics Joint Lab of Embodied AI, School of Computer Science, Peking University · AI2 Robotics · State Key Laboratory of Mechanical System and Vibration, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai, and also with Shanghai Key Laboratory of Intelligent Robotics, Shanghai Jiao Tong University, Shanghai

  4. DisFace3DNet: Explainable Facial Attractiveness Prediction via 3D Component Disentanglement

    Oct 8, 2026Fenggui Rao, Yan Luximon, Jie ZhangInterpretable MLRepresentation Disentanglement

    School of Design, The Hong Kong Polytechnic University, Hong Kong, China · School of Design, The Hong Kong Polytechnic University, Hong Kong, China, and the Laboratory for Artificial Intelligence in Design, Hong Kong Science Park, Hong Kong, China · Faculty of Applied Sciences, Macao Polytechnic University, Macau, China

  5. Representation Disentanglement for Fair Chest X-Ray Diagnosis

    Oct 5, 2026Yujie Sun, Ruizhe Li, Xiaowu SunChest X-Ray ClassificationRepresentation Disentanglement

    Academy of Artificial Intelligence and Advanced Technology, Xi’an Jiaotong-Liverpool University, China · Nottingham Biomedical Research Centre, School of Medicine, University of Nottingham, UK

  6. 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

  7. Revisiting Cross-Reconstruction for Generalizable Deepfake Detection

    Oct 1, 2026Bingjian Yang, Shilei Zhao, Zheng WangImage Forgery DetectionDigital Image Forensics

    National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, China

  8. What Builds the Scene? Luminance Dominates Geometry Formation in 3D Gaussian Splatting

    Sep 30, 2026Rezvan Joshaghani, Steven Cutchin3D Gaussian Splatting3D Reconstruction

    Boise State University

  9. Reason in Style: Discovering and Controlling Style in Language Models

    Sep 30, 2026Ioana Marinescu, Eric Karl Oermann, Kyunghyun ChoUnsupervised LearningLanguage Model Steering

    NYU · NYU & NYU Langone Health

  10. SafeStyle: Calibrated Style Residual Injection for Controllable Style-Leakage Trade-off in Diffusion Stylization

    Sep 18, 2026Zhangping Yang, Min Li, Song Yan +4Image Style TransferDiffusion-Based Image Editing

    Xi’an High-tech Research Institute, Shaanxi, China · University of Science and Technology of China, Anhui, China · Xi’an University of Architecture and Technology, Shaanxi, China

  11. Do speech foundation models really learn words?

    Sep 9, 2026Robin Huo, Ewan DunbarRepresentation LearningSpeech Foundation Models

    University of Toronto, Canada

  12. A Dominant Diffuse Phase in the Sparse Autoencoder Phase Diagram

    Sep 9, 2026Alexis D. PlascenciaFeature SuperpositionRepresentation Disentanglement

  13. High-probability guarantees for linear accessibility in feature superposition

    Sep 9, 2026Enrico VompaFeature SuperpositionNeural Representation Geometry

    Applied Artificial Intelligence Group Tallinn University of Technology, Estonia

  14. Certified Topological Interaction in Neural Representations: Exact Tests and the Statistic They Require

    Sep 8, 2026Sushovan MajhiDisentangled Representation LearningNeural Representation Geometry

    Data Science Program, George Washington University, Washington, D.C., USA.

  15. Reachability Is Not Generalization: Understanding Verb--Noun Decomposition in Assembly Action Recognition

    Sep 4, 2026Changyi Li, Yu XiaoCompositional GeneralizationHuman Activity Recognition

    Department of Information and Communications Engineering Aalto University 02150 Espoo, Finland

  16. Entangled Representations Amplify Collateral Damage in Unlearning

    Sep 2, 2026Evžen Wybitul, Tim G. J. Rudner, Christian Schroeder de WittMachine UnlearningRepresentation Disentanglement

    University of Oxford · University of Toronto

  17. A Deeper Analysis of Block-Sparse Featurizers

    Aug 27, 2026Alexandru-Iulius Jerpelea, Amith AnanthramSparse AutoencodersRepresentation Disentanglement

    Columbia University

  18. Defending Wearable VLMs Against Private Attribute Inference

    Aug 26, 2026Zhimin Li, Pan Wang, Jingxian Chen +4Privacy Protection in VLMsPrivacy-Utility Tradeoff

    Swanson School of Engineering, University of Pittsburgh · North Allegheny Senior High · College of Engineering, Michigan State University +1

  19. LightAIR: Lightweight Action Inversion and Riemannian Rectification for Text-based Person Anomaly Search

    Aug 10, 2026Yulun Zhang, Zixu Li, Zhiwei Chen +6Text-Based Person SearchRepresentation Disentanglement

    Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences Shenzhen, China · Shandong University Jinan, China

  20. Support Operation Factorization: Compositional Readout of Frozen Vision Encoders under Controlled Interventions

    Aug 6, 2026Zhongyao Wang, Wanli Ouyang, Taoyong Cui +1Visual Representation LearningRepresentation Disentanglement

    Fudan University · MMLab, The Chinese University of Hong Kong · Department of Computer Science and Engineering, The Chinese University of Hong Kong

  21. MIDAS: Mutual Information Disentanglement with Uncertainty-Aware Fusion for Incomplete Multimodal Sentiment Analysis

    Aug 6, 2026Yuhua Wen, Yingying Zhou, Qifei Li +4Missing-Modality LearningMultimodal Disentangled Representation Learning

    School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China · Zhongguancun Academy, Beijing 100094, China · Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing 100084, China +1

  22. QuerySplat: Decoupling Geometry and Appearance Representations in 3DGS Prediction

    Aug 2, 2026Yinglong Li, Donghui Shen, Xiaoyu Zhang +53D Gaussian SplattingRepresentation Disentanglement

    State Key Laboratory of Virtual Reality Technology and Systems, Beihang University · InSpatio Research · State Key Lab of CAD&CG, Zhejiang University

  23. What to Remove, What to Preserve: Dual-Ambiguity Rectification for All-in-One Image Restoration

    Jul 30, 2026Cencen Liu, Wen Yin, Dongyang Zhang +6Representation LearningRepresentation Disentanglement

    University of Electronic Science and Technology of China · University of Electronic Science and Technology of China, Chengdu, Sichuan, China · Jiigan Technology +1

  24. Controlling Embedding Spaces with Text-Conditioned Transformations

    Jul 24, 2026Joseph Fioresi, Fabian Caba Heilbron, Pankaj Nathani +2Representation DisentanglementMultimodal Embedding

    Institute of Artificial Intelligence, University of Central Florida, USA · ⋆Majority of work done as an intern at Adobe Research · Adobe Research, USA

  25. Scaling Interpretable Transformers with Parity Bottleneck Layers

    Jul 22, 2026Andrew Mack, Kraig Yuheng Tou, Mark Henry +2Feature SuperpositionTransformer Interpretability

    Principles of Intelligence · Independent

  26. Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence

    Jul 20, 2026Katarzyna Filus, Sebastian PokucińskiRepresentation DisentanglementSparse Autoencoders

    Institute of Theoretical and Applied Informatics, Polish Academy of Sciences, Bałtycka 5, 44-100 Gliwice, Poland · Department of Applied Informatics, Silesian University of Technology, Akademicka 16, 44-100 Gliwice, Poland

  27. Coordinated Disentanglement with Iterative Mode Discovery Under Hidden Correlations

    Jul 19, 2026Rong Hu, Ling ChenDisentangled Representation LearningLatent Variable Models

    State Key Laboratory of Blockchain and Data Security, Zhejiang University, Hangzhou, China · College of Computer Science and Technology, Zhejiang University, Hangzhou, China

  28. MultiLoReFT: Decoupling Shared and Modality-Specific Subspaces in Multimodal Learning via Low-Rank Representation Fine-Tuning

    Jul 18, 2026Sana Tonekaboni, Viktoria Schuster, Caroline UhlerCross-Modal Representation LearningMultimodal Disentangled Representation Learning

    Massachusetts Institute of Technology, Cambridge, MA, USA · Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, MA, USA · Vector Institute, Toronto, Canada +1

  29. OFD-Net: Teacher-Free Reliable Semi-supervised Medical Image Segmentation with Orthogonal Feature Disentanglement Net of Foreground-Background

    Jul 18, 2026Shao-feng Jiang, Zhe-yang Jing, Qin Lu +4Representation DisentanglementSemi-Supervised Medical Image Segmentation

    Department of Biomedical Engineering, and the Key Laboratory of Nondestructive Testing, Ministry of Education, Nanchang Hangkong University, Nanchang 330063, China

  30. OrDA: Orthogonal Disentanglement of Access Habits Framework for Homepage Marketing Block Recommendations

    Jul 15, 2026Lingxiao Zhang, Xiaobo Li, Tao XuRepresentation DisentanglementRecommender Systems

    Ant Group · Hangzhou, Zhejiang, China