arXiv：cs.LG· Dingling Yao, Kahaan Gandhi, Valentin Duruisseaux, Boris Bonev, Francesco Locatello, Anima Anandkumar·· 3 小时前AI 评分37
ReRoute：无需受控实验的科学仿真器反事实预测框架
Counterfactual Predictions in Scientific Emulators Without Controlled Experiments
AI 导读
研究人员提出 ReRoute 框架，将事实数据与部分机制知识结合，无需受控干预数据即可实现科学 what-if 预测。在气候仿真任务中，ReRoute 在严重 CO₂ 分布偏移下将聚合气候误差降低 18.2-31.8%，同时保持标准条件下的预测能力，成本仅为在额外受控仿真数据上重训练的很小一部分。
正文
Abstract:Many scientific questions require reasoning about what was never observed: What if the conditions, interventions, or history had been different? Models can predict accurately on observed data yet fail on such what-if queries when correlated inputs are varied independently. A common remedy is to add controlled simulation data in which these factors are explicitly disentangled, but this requires access to a simulator, can be computationally expensive, and inherits the simulator's modeling assumptions. We introduce ReRoute, a framework for targeted scientific what-if prediction that combines factual data with partial mechanistic knowledge, without requiring controlled intervention data for adaptation. ReRoute fixes the queried input of a pretrained backbone to a reference value, reintroduces its variation through a known mechanistic pathway, and fine-tunes on the original factual data, while leaving downstream effects to the learned dynamics. We provide a causal identification result for this construction under explicit structural assumptions, with the core argument machine-checked in Lean. After showing that ReRoute achieves highly accurate counterfactual predictions in a controlled advection-diffusion system where exact responses are available, we turn to state-of-the-art climate emulation. On held-out coupled-climate interventions, ReRoute reduces aggregate climate error by 18.2-31.8% under severe CO$_2$ distribution shifts while preserving skill under standard conditions, at a small fraction of the cost of retraining on additional controlled simulations, without even accounting for the substantial expense of generating such data. Finally, on an emulator trained from historical ERA5 reanalysis, where no counterfactual reference exists, ReRoute preserves substantially more of the surface warming implied by the observed boundary conditions under a fixed-CO$_2$ counterfactual.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML) |
| Cite as: | arXiv:2610.02252 [cs.LG] |
| (or arXiv:2610.02252v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02252 arXiv-issued DOI via DataCite |
Submission history
From: Dingling Yao [view email]
[v1]
Wed, 30 Sep 2026 19:16:07 UTC (4,081 KB)
来源：arXiv：cs.LG · arxiv.org