Unverified50% confidenceFactExact time
On LEAPS's pattern learning tasks, the longer the input, the worse the model performs.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/20/2026
First Seen
Valid until: 1/18/2027
Sources
Related Entities
Related Claims
UnverifiedExperiments show that after only a few generations of self-training, the variance of model outputs drops sharply and the text tends toward homogeneity78% similarVerifiedThe closer an event is to the cutoff date, the smaller its share in the training corpus, and the less clearly the model has mastered it.77% similarUnverifiedThe larger the context window, the more likely model performance is to degrade—the more information fed to the model, the worse it may perform.77% similarUnverifiedUnder a local quadratic model, a larger batch means fewer but longer update steps, with each step incurring a greater loss on curvature, resulting in a critical batch size that depends on the learning rate, beyond which sample efficiency declines.77% similarUnverifiedDuplicate content in training data can cause model memorization and performance degradation.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074005API
curl https://kongchang.com/api/v1/knowledge/claims/1074005MCP
get_claim(id=1074005)