Results for "timing labels"

Aligns transcripts with audio timestamps.

Learning from data by constructing “pseudo-labels” (e.g., next-token prediction, masked modeling) without manual annotation.

Minimizing average loss on training data; can overfit when data is limited or biased.

Human or automated process of assigning targets; quality, consistency, and guidelines matter heavily.

A structured collection of examples used to train/evaluate models; quality, bias, and coverage often dominate outcomes.

A scalar measure optimized during training, typically expected loss over data, sometimes with regularization terms.

Updating a pretrained model’s weights on task-specific data to improve performance or adapt style/behavior.

Systematic differences in model outcomes across groups; arises from data, labels, and deployment context.

Selecting the most informative samples to label (e.g., uncertainty sampling) to reduce labeling cost.

Training a smaller “student” model to mimic a larger “teacher,” often improving efficiency while retaining performance.

Assigning labels per pixel (semantic) or per instance (instance segmentation) to map object boundaries.

Measures a model’s ability to fit random noise; used to bound generalization error.

Measures divergence between true and predicted probability distributions.

Models that learn to generate samples resembling training data.

Assigning category labels to images.

Pixel-wise classification of image regions.

Using limited human feedback to guide large models.