Leviathan, Y.; Kalman, M.; and Matias, Y. 2023. Fast Inference from Transformers via Speculative Decoding. In Krause, A.; Brunskill, E.; Cho, K.; Engelhardt, B.; Sabato, S.; and Scarlett, J., eds., Proceedings of the 40th International Conference on Machine Learning, volume 202 of Proceedings of Machine Learning Research, 19274–19286. PMLR.Chen, C.; Borgeaud, S.; Irving, G.; Lespiau, J.-B.; Sifre, L.; and Jumper, J. 2023. Accelerating Large Language Model Decoding with Speculative Sampling. arXiv preprint arXiv:2302.01318.Angelopoulos, A. N.; and Bates, S. 2023. A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification. Foundations and Trends in Machine Learning, 16(4): 494–591.Angelopoulos, A. N.; Bates, S.; Fisch, A.; Lei, L.; and Schuster, T. 2024. Conformal Risk Control. In International Conference on Learning Representations (ICLR).Geifman, Y.; and El-Yaniv, R. 2017. Selective Classification for Deep Neural Networks. In Advances in Neural Information Processing Systems (NeurIPS).Mozannar, H.; and Sontag, D. 2020. Consistent Estimators for Learning to Defer to an Expert. In International Conference on Machine Learning (ICML).Alshiekh, M.; Bloem, R.; Ehlers, R.; Koenighofer, B.; Niekum, S.; and Topcu, U. 2018. Safe Reinforcement Learning via Shielding. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1).这篇论文最有意思的地方,不是又造了一个“更聪明”的模型,而是把一个老问题说透了:AI会提案,不代表AI能拍板;AI会算,不代表AI能担责。在硬约束序列决策里,真正值钱的不是“能不能跑”,而是“能不能在不翻车的前提下少跑几次昂贵求解器”。CGPA把这件事做成了一个可认证的合同,思路很工程,也很残酷:不信任AI,但允许它帮忙。图1就是整篇论文的骨架。一个不可信的提案源先给出一段长度为K的动作前缀,可信核心只做三件事:逐步验证是否越界、判断这段前缀值不值、不行就交给回退求解器兜底。这套结构的妙处在于,安全不依赖AI,收益却可以吃到AI的红利。讲人话就是:让AI负责“写草稿”,让可信系统负责“盖章”。