**摘要**
Calibration aligns a model's predictive uncertainty with the frequencies of its empirical outcomes and is important for understanding and trusting reported probabilities. Recent work shows that enforcing calibration at the level of individual predictors can improve ensemble accuracy and calibration, with mixture-of-experts (MoE) models showing strong empirical improvements in particular; however,
👤 作者: Gina Wong, Drew Prinster, Suchi Saria, Rama Chellappa, Anqi Liu

---
🔗 **[Toward Calibrated Mixture-of-Experts Under Distribution Shift](https://arxiv.org/abs/2606.20544v1)**

> Toward Calibrated Mixture-of-Experts Under Distribution Shift
🏷️ 来源: ArXiv cs.AI
⏱️ 2026-06-19 22:01