Figure 1 - Bayes' Power for Explaining In-Context Learning Generalizations

Figure 1: The model is only trained on step functions (left), still it learns to make smooth predictions (right) just like the true posterior for the step function prior.

The model is only trained on step functions (left), still it learns to make smooth predictions (right) just like the true posterior for the step function prior.

Go to figure page

Reference

Bayes' Power for Explaining In-Context Learning Generalizations - Scientific Figure on ResearchGate. Available from: https://www.researchgate.net/figure/The-model-is-only-trained-on-step-functions-left-still-it-learns-to-make-smooth_fig1_384597999 [accessed 24 Apr 2025]

Caption

Figure 1: The model is only trained on step functions (left), still it learns to make smooth predictions (right) just like the true posterior for the step function prior.

Embed code

<a href="https://www.researchgate.net/figure/The-model-is-only-trained-on-step-functions-left-still-it-learns-to-make-smooth_fig1_384597999"><img src="https://www.researchgate.net/publication/384597999/figure/fig1/AS:11431281281697077@1727968872681/The-model-is-only-trained-on-step-functions-left-still-it-learns-to-make-smooth.png" alt="The model is only trained on step functions (left), still it learns to make smooth predictions (right) just like the true posterior for the step function prior."/></a>