The story’s score and claims

Comments on No. 4 of 62 · arXiv · edition of Sunday 15 November 2026

The Curse of Recursion: Training on Generated Data Makes Models Forget (2024)

Summary

Researchers at Cambridge, Oxford, Imperial, Toronto and Edinburgh show that models trained recursively on their predecessors' output lose the tails of the original distribution and drift toward a narrow point estimate. They demonstrate this in Gaussian mixtures, variational autoencoders and fine-tuned language models, and argue from finite sampling error that it is unavoidable. For teams that train on scraped web data, human-generated data and provenance become more valuable. The language-model results come from fine-tuning, where only early signs appear, and several quoted figures lacked sources.

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules