Conformalized graph neural network emulation of regional climate extremes: an artificial intelligence evaluation of coverage guarantees under emissions-scenario shift
Oct 2026· Zenodo (CERN European Organization for Nuclear Research)
Climate variability and models
Abstract
Conformal prediction offers distribution-free coverage guarantees for machine-learning emulators of climate extremes, but the guarantee is only as good as the exchangeability assumption that underwrites it. Using NASA POWER monthly reanalysis for Kampala, Uganda (0.35°N, 32.58°E) over 2000–2023 (312 observations, mean 22.443 °C, linear trend −0.1564 °C per decade), the study specifies a graph construction, a split-conformal calibration protocol, and a coverage-degradation bound expressed as a function of a scenario-shift divergence. The analysis derives a finite-sample coverage gap under covariate shift, proposes a weighted conformal score that restores approximate validity, and identifies the conditions under which GNN emulators of regional extremes can be certified for out-of-distribution emissions scenarios. The contribution is a formal account of when conformal coverage survives scenario shift and a diagnostic that practitioners can compute before deploying an emulator.
This paper investigates developing real-world perception for active agents, proposes Gibson Environment for this purpose, and showcases a set of perceptual tasks learned therein.
F. Xia, Amir Zamir, Zhi-Yang He et al.· 2018 IEEE/CVF Conference on...· 1,069 citations· ⚡112
GAOKAO-Bench is introduced, an intuitive benchmark that employs questions from the Chinese GAOKAO examination as test samples, including both subjective and objective questions that contribute a robust evaluation benchmark for future large language models and offers valuable insights into the advantages and limitations...
Xiaotian Zhang, Chun-yan Li, Yi Zong et al.· arXiv.org· 216 citations· ⚡17
An algorithm is proposed which leverages disentangled semantic factors to generate adversarial perturbation by altering controlled semantic attributes to fool the learner towards various "adversarial" targets.
Haonan Qiu, Chaowei Xiao, Lei Yang et al.· arXiv.org· 208 citations· ⚡17
This work proposes a HALO that directly maximizes the utility of generations instead of maximizing the log-likelihood of preferences, and suggests that the best loss depends on the inductive biases most appropriate for a given setting, an oft-overlooked consideration.
This work investigates the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection and finds that the MTEs are biased, significantly favoring White-associated names in 85% of cases and female-associated names in only 11.1% of cases.
This work shows that orders of magnitude enhancement in performance could be obtained by a combination of hardware improvements and tight quantum-HPC integration and introduces high-performance architectures for quantum-probabilistic computing with custom-designed accelerators to tackle today's industry-scale classical...
Masoud Mohseni, Artur Scherer, K. Johnson et al.· arXiv.org· 121 citations· ⚡9
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 13, 2026
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
