mlsanityData to 7 Oct 2026

Independent Density Estimation

arXiv 2512.10067 ↗First posted 10 Dec 2025cs.CV cs.LGcited by 0

On the map: Multimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP

Abstract (arXiv)

Large-scale Vision-Language models have achieved remarkable results in various domains, such as image captioning and conditioned image generation. Nevertheless, these models still encounter difficulties in achieving human-like compositional generalization. In this study, we propose a new method called Independent Density Estimation (IDE) to tackle this challenge. IDE aims to learn the connection between individual words in a sentence and the corresponding features in an image, enabling compositional generalization. We build two models based on the philosophy of IDE. The first one utilizes fully disentangled visual representations as input, and the second leverages a Variational Auto-Encoder to obtain partially disentangled features from raw images. Additionally, we propose an entropy-based compositional inference method to combine predictions of each word in the sentence. Our models exhibit superior generalization to unseen compositions compared to current models when evaluated on various datasets.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 18 of its 23 references that are in the map. First posted on arXiv 10 Dec 2025.

Builds on

18 of 23 references.

References · 18

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2021TransUNet: Transformers Make Strong Encoders for Medical Image Segmentationcited by 1,552
  6. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,445
  7. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 850
  8. 2021Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigmcited by 405
  9. 2019The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervisioncited by 369
  10. 2020Keystroke Dynamics as Part of Lifeloggingcited by 334
  11. 2018Towards a Definition of Disentangled Representationscited by 294
  12. 2018BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learningcited by 238
6 more
  1. 2019Scaling Autoregressive Video Modelscited by 189
  2. 2019Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEscited by 116
  3. 2022Does CLIP Bind Concepts? Probing Compositionality in Large Image Modelscited by 101
  4. 2021CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIPcited by 88
  5. 2020Visual Concept-Metaconcept Learningcited by 39
  6. 2020Compositional Networks Enable Systematic Generalization for Grounded Language Understandingcited by 18

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Language Plays a Pivotal Role in the Object-Attribute Compositional Generalization of CLIP
  2. 2023Towards Efficient Vision-Language Tuning: More Information Density, More Generalizability
  3. 2026Gate-and-Merge: Zero-shot Compositional Personalization of Vision Language Models
  4. 2022CREPE: Can Vision-Language Foundation Models Reason Compositionally?
  5. 2023SugarCrepe: Fixing Hackable Benchmarks for Vision-Language Compositionality
  6. 2024The ART of Composition: Attention-Regularized Training for Compositional Visual Grounding
  7. 2024Compositional Entailment Learning for Hyperbolic Vision-Language Models
  8. 2025Evaluating Compositional Generalisation in VLMs and Diffusion Models