Results for "data → model"

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

Raw model outputs before converting to probabilities; manipulated during decoding and calibration.

System for running consistent evaluations across tasks, versions, prompts, and model settings.

Inputs crafted to cause model errors or unsafe behavior, often imperceptible in vision or subtle in text.

System design where humans validate or guide model outputs, especially for high-stakes decisions.

Constraining model outputs into a schema used to call external APIs/tools safely and deterministically.

A model is PAC-learnable if it can, with high probability, learn an approximately correct hypothesis from finite samples.

A measure of a model class’s expressive capacity based on its ability to shatter datasets.

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

Using same parameters across different parts of a model.

The range of functions a model can represent.

Allows model to attend to information from different subspaces simultaneously.

Capabilities that appear only beyond certain model sizes.

Logged record of model inputs, outputs, and decisions.

Probabilistic graphical model for structured prediction.

Graphical model expressing factorization of a probability distribution.

Model that compresses input into latent space and reconstructs it.

Joint vision-language model aligning images and text.

Classical statistical time-series model.

End-to-end process for model training.

Model execution path in production.

Running new model alongside production without user impact.

Shift in model outputs.

Increasing model capacity via compute.

Cost of model training.

Model exploits poorly specified objectives.

Model optimizes objectives misaligned with human values.

Model behaves well during training but not deployment.

Assigning a role or identity to the model.

Asking model to review and improve output.