accuracy gap
The difference in performance metrics between two models or systems, often used to compare the effectiveness of diverse methods in achieving predictive accuracy, with implications for model selection and interpretability.
- Competitive Advantage Attacks to Decentralized Federated Learning
- Neural Tangent Knowledge Distillation for Optical Convolutional Networks
- Same Task, Different Circuits: Disentangling Modality-Specific Mechanisms in VLMs
- SoMi-ToM: Evaluating Multi-Perspective Theory of Mind in Embodied Social Interactions