compositional generalization
Compositional generalization is the ability of a model to understand and generate new combinations of known components or concepts. It's crucial for achieving flexible and robust reasoning in AI systems, especially in natural language processing and computer vision.
- Bridging the gap to real-world language-grounded visual concept learning
- Dyn-O: Building Structured World Models with Object-Centric Representations
- In-Context Compositional Learning vis Sparse Coding Transformer
- Role Bias in Diffusion Models: Diagnosing and Mitigating through Intermediate Decomposition
- RrED: Black-box Unsupervised Domain Adaptation via Rectifying-reasoning Errors of Diffusion
- Scalable Evaluation and Neural Models for Compositional Generalization
- Scaling Computer-Use Grounding via User Interface Decomposition and Synthesis
- Scaling can lead to compositional generalization
- Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following
- Unveiling the Compositional Ability Gap in Vision-Language Reasoning Model
- Volume Transmission Implements Context Factorization to Target Online Credit Assignment and Enable Compositional Generalization