Metraipolis › archive › dc_8a4e14db4fd8
The Delta Gap: Cellular World Models for Perturbation Response Prediction Are Far Harder Than Global Correlation Implies — arXiv:2611.10658
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.10658
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-10T18:41:08.012Z
- SHA-256 of the archived text
632e69c78877ba284ee79cad7dde10aee352b6054f076d021941832f299b14ff- Stable URL of this page
https://metraipolis.com/documento/dc_8a4e14db4fd8
How to cite
The Delta Gap: Cellular World Models for Perturbation Response Prediction Are Far Harder Than Global Correlation Implies — arXiv:2611.10658 — arXiv (en), https://arxiv.org/abs/2611.10658 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8a4e14db4fd8 (Metraipolis archive dc_8a4e14db4fd8, sha256 632e69c78877ba284ee79cad7dde10aee352b6054f076d021941832f299b14ff)
BibTeX
@misc{dc8a4e14db4fd8,
title = {The Delta Gap: Cellular World Models for Perturbation Response Prediction Are Far Harder Than Global Correlation Implies — arXiv:2611.10658},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.10658},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8a4e14db4fd8; sha256 632e69c78877ba284ee79cad7dde10aee352b6054f076d021941832f299b14ff},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10658. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Cellular world models aim to predict how cells respond to genetic perturbations, analogous to world models in reinforcement learning. Perturbation response models are evaluated almost exclusively by global Pearson correlation between predicted and observed transcriptomes. We show this metric is dominated by thousands of non-responsive genes and severely inflates apparent performance. On the Replogle K562 screen (509 unseen perturbations, 5000 genes), GEARS achieves global Pearson 0.977, but its delta Pearson is only 0.076. A trivial baseline predicting mean training delta achieves 0.267, which is 3.5x better than GEARS. Our primary contribution is methodological: we identify and quantify this delta gap as a systematic evaluation artifact, and propose delta-space evaluation as the primary benchmark. We provide two post-hoc corrections (RWR on co-expression graphs and kNN matching) and show that 2.8x of the remaining gap lies in perturbation representation geometry, not decoder capacity. We further show the gap is partially addressable end-to-end: delta-aligned training (lambda = 3.0) raises delta Pearson to 0.179 on Replogle (2.35x over GEARS) and surpasses the trivial baseline on direction accuracy (0.680 vs 0.567) and gene recovery (1.52 vs 1.19); combined with kNN it reaches 0.217. An amplitude-calibration variant (eta = 0.3) further raises gene recovery to 2.16 while reaching 0.218. Critically, these representations transfer zero-shot across cell lines: from Replogle K562 to unseen Norman A549, the delta-aligned model reaches 0.328 on 741 shared genes, 4.7x a trivial baseline and 2.1x standard GEARS. This is the first systematic quantification of the delta gap, and the first diagnostic framework localizing the residual gap to perturbation representation geometry via an oracle upper bound. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.10658 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-10T18:37:37.498Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).