Classifier
A model in machine learning that maps inputs to one of a finite set of predefined categories, outputting a probability distribution over those categories. The candidate space is fixed and closed, distinguishing classifiers from generative language models, which perform open-vocabulary prediction.
Timeline (last 90 days)
Defenses against prompt injection (classifiers, instruction hierarchies, source labels) are without exception probabilistic; the model may be bypassed by novel phrasing, and classifiers may misjudge legitimate input.
A classifier maps input to one of a finite set of predefined categories with a fixed, closed candidate space; a generative language model predicts the next token from the entire vocabulary at each step, with an open candidate space, building output autoregressively.
Detectors are responsible for locating targets in images or videos and outputting bounding boxes and confidence scores, while classifiers determine the species or category of the target.
Classification tasks such as sentiment analysis, spam filtering, intent recognition, and content tagging do not require the complex reasoning capabilities of generative large models, making lightweight classifiers more efficient and controllable.
Locally-run classifiers can operate on ordinary laptops, servers, or even edge devices without relying on GPU resources or cloud GPU instances.
The classifier's default decision threshold systematically inflates salience estimates at the corpus level.
All Facts (6)
Defenses against prompt injection (classifiers, instruction hierarchies, source labels) are without exception probabilistic; the model may be bypassed by novel phrasing, and classifiers may misjudge legitimate input.
50%UnverifiedA classifier maps input to one of a finite set of predefined categories with a fixed, closed candidate space; a generative language model predicts the next token from the entire vocabulary at each step, with an open candidate space, building output autoregressively.
50%UnverifiedDetectors are responsible for locating targets in images or videos and outputting bounding boxes and confidence scores, while classifiers determine the species or category of the target.
50%UnverifiedClassification tasks such as sentiment analysis, spam filtering, intent recognition, and content tagging do not require the complex reasoning capabilities of generative large models, making lightweight classifiers more efficient and controllable.
50%UnverifiedLocally-run classifiers can operate on ordinary laptops, servers, or even edge devices without relying on GPU resources or cloud GPU instances.
50%UnverifiedThe classifier's default decision threshold systematically inflates salience estimates at the corpus level.
50%