Compositional Representation Learning

Overview

The field studies how models represent reusable concepts, attributes, objects, and relations so they can combine them in new ways. Approaches include factorized and tensor-product representations, object-centric or graph-based models, sparse autoencoders, and contrastive learning for vision-language models. Researchers test these ideas on compositional zero-shot recognition, image-text matching benchmarks such as SugarCrepe, and controlled tasks such as CLEVR, while probing learned features and their interactions. A central tension is that strong performance on unseen combinations does not by itself show that a model learned reusable structure: shortcuts and dataset biases can also produce apparent generalization.

Latest papers

CardsList

97 papers17 papers in the last four weeks, down 35% on the four weeks before. 0.1% of all new papers.

  1. PRISM: Primitive-Integrated Separable Modulation with Priority Visual Memory Queue for Compositional Zero-Shot Learning

    Nov 15, 2026Jianhua Wang, Zhihang Zheng, Yang Liu +1Zero-Shot LearningTest-Time Adaptation of VLMs

  2. Linking Rapid Learning to Compositional Structure in Neural Networks: Evidence from Additive Analogies in Meta-Learned Parameter Spaces

    Nov 12, 2026Abigail L. Tenenbaum, R. Thomas McCoyMeta-LearningCompositional Representation Learning

  3. Beyond Row-Level Prediction: Evaluating and Learning Composition for Reusable Table Representations

    Nov 10, 2026Ali Younes, Saeed Ghoorchian, Maximilian Schambach +2Tabular Representation LearningCompositional Representation Learning

  4. Chem World: A Large-Scale Benchmark and Physics-Aware Framework for Trustworthy Molecular Property Prediction

    Nov 10, 2026Tianyou Bai, Huan Wang, Mingchen Gao +5Molecular Property PredictionCompositional Representation Learning

  5. Factorized MLLM Embeddings for Composed Image Retrieval

    Nov 3, 2026WonJun Moon, Seongchan Kim, Dahun Kim +1Composed Image RetrievalMultimodal Embedding

  6. DualClusterNet: Personalized Pattern-Space Composition for Next-POI Recommendation

    Nov 1, 2026Ziyang Su, Guanjie Zheng, Jilin Hu +1Sequential RecommendationUser Preference Modeling

  7. Efficient Learning of Mechanism Designs and Compositionality

    Oct 29, 2026João Pedro Gandarela, Thiago Rios, Stefan Menzel +1Compositional Representation LearningData-Efficient Learning

  8. TaxoMFM: A Taxonomy-Aware Foundation Model for the Human Gut Microbiome

    Oct 29, 2026Haochen Xue, Anastasia Godneva, Yulong Li +6BioinformaticsHierarchical Representation Learning

  9. Relational Composition as Parallel Transport: What Per-Entity Frames Force in Attention

    Oct 26, 2026Aakash Sarkar, Ananya Mukherjee, Motoshige SatoRelational ReasoningCompositional Representation Learning

  10. Compositionality with Primitives: A Unified View from Euclidean to Hyperbolic Vision-Language Models

    Oct 21, 2026Puru Vaish, Meelad Dashti, Christoph Brune +3Compositional Reasoning in VLMsZero-Shot Learning

  11. Learning Computational Structure for Sample-Efficient Prediction

    Oct 20, 2026Hangyu Zhou, Constantine Dovrolis, Stephen MussmannData-Efficient LearningCompositional Representation Learning

  12. LSA-PAST: Layer-Specific Adaptation and Prototype Transport for Compositional Incremental Learning

    Oct 19, 2026Kyle Stein, Andrew Arash Mahyari, Guillermo Francia +2Incremental LearningClass-Incremental Learning

  13. Flow-CLIP: Image-Conditioned Residual Scoring for CompositionalImage-Text Matching

    Oct 15, 2026Minghong Wei, Enguang Zhu, Pu Cao +2Vision-Language AlignmentCross-Modal Feature Matching

  14. Behavioral Compositionality of Sparse Autoencoder Representations

    Oct 15, 2026Haotian Ma, Ziye Mei, Frederic SalaSparse AutoencodersCompositional Representation Learning

  15. LLMs program using Reusable and Composable Abstract Representations

    Oct 13, 2026Casper L. Christensen, Kyle J. Ray, Dmitry Manning-Coe +2LLM Code GenerationCompositional Representation Learning

  16. Virtual-Alloy Physics-Aware Semantic System for extrapolative prediction of perovskite solar cells

    Oct 13, 2026Qiwei Han, Chi Zhouperovskite solar cellsMaterials Property Prediction

  17. MSVAE: Rethinking Musical Semantics in Continuous Latent Spaces

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

  18. Learning Perceptual Geometry of Natural Odor Mixtures from LLM-Derived Descriptors

    Oct 12, 2026Farhad Pashakhanloo, Venkatesh N. MurthyMolecular Representation LearningCompositional Representation Learning

  19. Binding Phoneme Identity to Position in Neural Models and the Human Brain

    Oct 11, 2026Nafiseh Momeni, Malo Renaudin, Itzhak Fried +1Neural EncodingNeural Representation Geometry

  20. CAFE+FNO: Fourier Kernel Generation via Multiplicative Feature Composition

    Oct 7, 2026Hyungjoon Juen, Minwoo ShinFourier Neural OperatorPDE Operator Learning

  21. From Global Alignment to Local Grounding: Zero-Shot Chinese Character Recognition with Radical Verification

    Oct 7, 2026Yu-Heng Shih, Bing-Chen Wu, Tsz-To Wong +6Zero-Shot LearningOptical Character Recognition

  22. SCION: Scene Composition with Instanced Neural Primitives

    Oct 1, 2026William Koch, Amogh Joshi, Cyrus Vachha +23DGS Compression3D Scene Editing

  23. Learning Expressive and Compositional Motion Representation via Spectral Skills

    Sep 29, 2026Feiyang Wu, Chenxiao Gao, Chen Yang +3Robot Skill LearningHumanoid Robot Control