Unverified50% confidenceFactExact time
Experiments show that a randomly initialized latent predictor and a latent predictor pretrained on screen dynamics perform comparably in GUI-JEPA, i.e., the latent predictor used for decision-making does not need pretraining.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe in-context predictor is optimal for topics never seen during training; it relies on the prompt for on-the-fly adaptation rather than on stored parameters.76% similarUnverifiedTraining from scratch means not using any pretrained weights, but instead completing the full training process on a new dataset with randomly initialized parameters.71% similarUnverifiedThe core prediction of this theory is: the in-context predictor is available very early, already existing when the amount of training data is still small; whereas the specialised predictor only emerges after the number of training tokens exceeds the number of parameters of the data generation process.71% similarUnverifiedThe pretraining and fine-tuning paradigm originated from NLP and computer vision, with its core being large-scale data pretraining to learn general low-level features, and downstream tasks adapted using a small amount of labeled data.70% similarUnverifiedDecision-focused learning (Predict-then-Optimize) advocates using downstream decision quality directly as the training signal.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1106120API
curl https://kongchang.com/api/v1/knowledge/claims/1106120MCP
get_claim(id=1106120)