Pith. sign in

Reference change · event page

Reference changes · DOI

AI models collapse when trained on recursively generated data.Nature, 631:755–759, 2024

Published notice on a work cited in the Pith corpus. Exact quotes below. No model judges whether any citation was load-bearing.

This page records that a citing paper's bibliography includes a work with a published notice. It is not a judgment on the citing paper.

Correction Crossref 28 open · 28 total · 0 disputed
DOI
10.1038/s41586-024-07566-y
Notice DOI
10.1038/s41586-025-08905-3
Event date
2025-03-21
Machine twin
JSON

01One-hop citing occurrences

Correction Open
Iterative Finetuning is Mostly Idempotent

ref [10] · 2605.01130 · notice #987 · dispute

Raw extraction · bibliography line

ISSN 1476-4687. doi: 10.1038/s41586-024-07566-y. Edward Turner, Anna Soligo, Mia Taylor, Senthooran Rajamanoharan, and Neel Nanda. Model organisms for emergent misalignment,

Parser render (TeX stripped for reading; raw above is the evidence)

ISSN 1476-4687. doi: 10.1038/s41586-024-07566-y. Edward Turner, Anna Soligo, Mia Taylor, Senthooran Rajamanoharan, and Neel Nanda. Model organisms for emergent misalignment

Correction Open
LLMs Have Made Failure Worth Publishing

ref [10] · 2604.06236 · notice #992 · dispute

Raw extraction · bibliography line

doi:10.1038/s41586-024-07566-y. Alessandra Toniato, Alain C. Vaucher, Teodoro Laino, and Mara Graziani. Negative chemical data boosts language models in reaction outcome prediction.Science Advances, 11(24),

Parser render (TeX stripped for reading; raw above is the evidence)

doi:10.1038/s41586-024-07566-y. Alessandra Toniato, Alain C. Vaucher, Teodoro Laino, and Mara Graziani. Negative chemical data boosts language models in reaction outcome prediction.Science Advances, 11(24)

Correction Open
Model Collapse as Cultural Evolution

ref [3] · 2605.23054 · notice #1001 · dispute

Raw extraction · bibliography line

Ilia Shumailov and Zakhar Shumaylov and Yiren Zhao and Nicolas Papernot and Ross J. Anderson and Yarin Gal , title =. Nature , volume =. 2024 , burl =. doi:10.1038/S41586-024-07566-Y , timestamp =

Parser render (TeX stripped for reading; raw above is the evidence)

Ilia Shumailov and Zakhar Shumaylov and Yiren Zhao and Nicolas Papernot and Ross J. Anderson and Yarin Gal, title =. Nature, volume =. 2024, burl =. doi:10.1038/S41586-024-07566-Y, timestamp =

Correction Open
q0: Primitives for Hyper-Epoch Pretraining

ref [19] · 2606.03938 · notice #1004 · dispute

Raw extraction · bibliography line

Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao, Nicolas Papernot, Ross Anderson, and Yarin Gal. AI models collapse when trained on recursively generated data.Nature, 631(8022):755– 759, 2024. doi: 10.1038/s41586-024-07566-y
Correction Open
Hierarchical Synthetic Tabular Data Generation: A Hybrid Top-Down and Bottom-Up Framework

ref [1] · 2605.28198 · notice #1007 · dispute

Raw extraction · bibliography line

doi: 10.1038/s41586-024-07566-y. URL https: //doi.org/10.1038/s41586-024-07566-y. Solatorio, A. V . and Dupriez, O. Realtabformer: Generating realistic relational and tabular data using transformers. arXiv preprint arXiv:2302.02041, 2023. Turanyksel, S. Adult income dataset. https://www. kaggle.com/datasets/serpilturanyksel/ adult-income, 2021. Umesh, C., Seegel-Schultz, K., Mahendra, M., Bej, S., and Wolkenhauer, O. Dependency-aware synthetic tabular data generation.Pattern Recog- nition, 179:113819, 2026. ISSN 0031-3203. doi: https://doi.org/10.1016/j.patcog.2026.113819. URL https://www.sciencedirect.com/ science/article/pii/S0031320326007843. Wang, Y ., Feng, D., Dai, Y ., Chen, Z., Huang, J., Anani- adou, S., Xie, Q., and Wang, H. Harmonic: Harnessing llms for tabular data synthesis and privacy protection. In Globerson, A., Mackey, L., Belgrave, D., Fan, A., Pa- quet, U., Tomczak, J., and Zhang, C. (eds.),Advances in Neural Information Processing Systems, volume 37, pp. 100196–100212. Curran Associates, Inc., 2024. doi: 10.52202/079017-3179. Xu, L., Skoularidou, M., Cuesta-Infante, A., and Veeramachaneni, K. Modeling tabular data using conditional gan. In Wallach, H., Larochell
Correction Open
When transformers learn "impossible" languages, what do they learn?

ref [3] · 2606.30815 · notice #1008 · dispute

Raw extraction · bibliography line

Shumailov, Ilia and Shumaylov, Zakhar and Zhao, Yiren and Papernot, Nicolas and Anderson, Ross and Gal, Yarin , month = jul, year =. Nature , publisher =. doi:10.1038/s41586-024-07566-y , abstract =

Parser render (TeX stripped for reading; raw above is the evidence)

Shumailov, Ilia and Shumaylov, Zakhar and Zhao, Yiren and Papernot, Nicolas and Anderson, Ross and Gal, Yarin, month = jul, year =. Nature, publisher =. doi:10.1038/s41586-024-07566-y, abstract =

Correction Open
Agentic AI Scientists Are Not Built For Autonomous Scientific Discovery

ref [2] · 2605.08956 · notice #991 · dispute

Raw extraction · bibliography line

ISSN 1476-4687. doi: 10.1038/s41586-024-07566-y. Joseph P. Simmons, Leif D. Nelson, and Uri Simonsohn. False-positive psychology: Undisclosed flexibility in data collection and analysis allows presenting anything as significant.Psychological Science, 22(11):1359–1366, 2011. ISSN 1467-9280. doi: 10.1177/0956797611417632. Leonid Teytelman, Alexei Stoliartchouk, Lori Kindler, and Bonnie L. Hurwitz. Protocols.io: Virtual Communities for Protocol Development and Discussion.PLOS Biology, 14(8):e1002538, August

Status lifecycle on notices: open → disputed → (response attached on the notice page). Repaired counts matter as much as open counts. There is no “safe,” “invalid,” or “resolved” badge.