arXiv AI
What Drives Compositional Generalization in Visual Generative Models? The Importance of Continuous Training Objectives
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Jul 7
Is Generation Required for Data-Efficient Perception?
arXiv:2512. 08854v3 Announce Type: replace-cross Abstract: It has been hypothesized that achieving the data efficiency of human visual perception requires a generative approach in which internal representations result from inverting a decoder.
arXiv AI
Jun 9
When Do Diffusion Models learn to Generate Multiple Objects?
arXiv:2605. 00273v2 Announce Type: replace-cross Abstract: Text-to-image diffusion models achieve impressive visual fidelity, yet they remain unreliable in multi-object generation.
arXiv AI
Jun 12
Cross-Modal Masked Compositional Concept Modeling for Enhancing Visio-Linguistic Compositionality
arXiv:2606. 13288v1 Announce Type: cross Abstract: Contrastively trained vision-language models like CLIP, have made remarkable progress in learning joint image-text representations, but still face challenges in compositional understanding.
arXiv Machine Learning
Jul 7
Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models
arXiv:2602. 24264v2 Announce Type: replace-cross Abstract: Compositional generalization, the ability to recognize familiar parts in novel contexts, is a defining property of intelligent systems.
arXiv Machine Learning
Jul 3
Optimizing Visual Generative Models via Distribution-wise Rewards
arXiv:2607. 02291v1 Announce Type: new Abstract: Conventional reinforcement learning strategies for visual generation typically employ sample-wise reward functions, yet this practice frequently results in reward hacking that degrades image diversity and introduces visual anomalies.
arXiv AI
Jun 2
Global Geometry Is Not Enough for Vision Representations
arXiv:2602. 03282v2 Announce Type: replace-cross Abstract: A common assumption in representation learning is that globally well-distributed embeddings support robust and generalizable representations.