mlsanityData to 14 Oct 2026

Learning Adaptive and Visually Aligned Conditions for Generative Zero-Shot Learning

arXiv 2603.06281 ↗First posted 6 Mar 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Generative zero-shot learning (ZSL) synthesizes visual features for unseen classes by learning a semantic-conditioned generator from seen classes. Since semantic conditions determine the knowledge transfer from seen to unseen classes, their effectiveness is critical for generating plausible features. However, existing class-level semantic prototypes lack semantic diversity to capture intra-class variations, while exhibiting inconsistent inter-class structures with the visual space due to the semantic-visual gap. To address these issues, we propose Adaptive Attribute Distribution and Visual Structure Alignment (AAVS), a unified semantic condition learning framework for generative ZSL. Specifically, the Adaptive Attribute Distribution (AAD) learns dimension-adaptive attribute distributions with discriminability-guided variation calibration to capture transferable semantic diversity, while Visual Structure Alignment (VSA) aligns the diverse attributes with refined visual prototypes to learn visual inter-class structures. By jointly modeling semantic diversity and visual structural consistency, AAVS provides more effective generation conditions for unseen classes to generate visual features. Extensive quantitative and qualitative evaluations on three benchmarks demonstrate that AAVS outperforms existing generative ZSL methods and validate the effectiveness of the learned generation conditions.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 23 of its 38 references that are in the map. First posted on arXiv 6 Mar 2026.

Builds on

23 of 38 references.

References · 23

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  3. 2014Conditional Generative Adversarial Netscited by 2,162
  4. 2017Zero-Shot Learning -- A Comprehensive Evaluation of the Good, the Bad and the Uglycited by 538
  5. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 516
  6. 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
  7. 2017Feature Generating Networks for Zero-Shot Learningcited by 257
  8. 2016Unrolled Generative Adversarial Networkscited by 208
  9. 2016Learning Deep Representations of Fine-grained Visual Descriptionscited by 182
  10. 2019f-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learningcited by 164
  11. 2016Learning a Deep Embedding Model for Zero-Shot Learningcited by 153
  12. 2019Seeing What a GAN Cannot Generatecited by 148
11 more
  1. 2016Latent Embeddings for Zero-shot Classificationcited by 129
  2. 2021FREE: Feature Refinement for Generalized Zero-Shot Learningcited by 67
  3. 2023Improving Zero-Shot Generalization for CLIP with Synthesized Promptscited by 51
  4. 2021TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learningcited by 28
  5. 2023Evolving Semantic Prototype Improves Generative Zero-Shot Learningcited by 20
  6. 2024Progressive Semantic-Guided Vision Transformer for Zero-Shot Learningcited by 17
  7. 2024Visual-Augmented Dynamic Semantic Prototype for Generative Zero-Shot Learningcited by 16
  8. 2024ZeroDiff: Solidified Visual-Semantic Correlation in Zero-Shot Learningcited by 10
  9. 2025GenZSL: Generative Zero-Shot Learning Via Inductive Variational Autoencodercited by 5
  10. 2025Visual and Semantic Prompt Collaboration for Generalized Zero-Shot Learningcited by 4
  11. 2024PSVMA+: Exploring Multi-granularity Semantic-visual Adaption for Generalized Zero-shot Learningcited by 3

Built on by

Not cited yet.