training-free approach
A training-free approach in AI refers to methods that require no or minimal learning from data. This could include rule-based systems or algorithms that exploit prior knowledge without requiring empirical training.
- CoFFT: Chain of Foresight-Focus Thought for Visual Language Models
- Foundation Cures Personalization: Improving Personalized Models’ Prompt Consistency via Hidden Foundation Knowledge
- Generative Model Inversion Through the Lens of the Manifold Hypothesis
- Instance-Level Composed Image Retrieval
- Same Task, Different Circuits: Disentangling Modality-Specific Mechanisms in VLMs
- Training-Free Safe Text Embedding Guidance for Text-to-Image Diffusion Models
- Training-free Detection of AI-generated images via Cropping Robustness
- Training-free Online Video Step Grounding
- Vision Transformers Don't Need Trained Registers
- Visual Jenga: Discovering Object Dependencies via Counterfactual Inpainting