Conformal Thinking：计算预算下推理的风险控制框架
阅读原文· machinelearning.apple.com推理大语言模型支持测试时扩展，准确率随 token 预算增加而提升，但预算设定带来风险-精度权衡。Conformal Thinking 框架将预算设定重定义为风险控制问题：在最小化计算量的同时限制错误率。该框架引入上阈值（模型足够自信时停止推理，承担输出错误的风险）和下阈值（提前终止无法解决的实例，承担过早停止的风险）。给定目标风险与验证集后，使用无分布风险控制来最优指定这些停止机制。跨多种推理任务和模型的实验表明，该方法在遵守用户指定风险目标的同时，通过下阈值与集成停止机制实现了计算效率提升。代码已开源。
Conformal Thinking: Risk Control for Reasoning on a Compute Budget
AuthorsXi Wang†*, Anushri Suresh†*, Alvin Zhang†*, Rishi More†*, William Jurayj†, Benjamin Van Durme†, Mehrdad Farajtabar, Daniel Khashabi†, Eric Nalisnick†
Reasoning Large Language Models (LLMs) enable test-time scaling, with dataset-level accuracy improving as the token budget increases, motivating adaptive reasoning—spending tokens when they improve reliability and stopping early when additional computation is unlikely to help. However, setting the token budget, as well as the threshold for adaptive reasoning, is a practical challenge that entails a fundamental risk-accuracy trade-off. We re-frame the budget setting problem as risk control, limiting the error rate while minimizing compute. Our framework introduces an upper threshold that stops reasoning when the model is confident (risking incorrect output) and a novel parametric lower threshold that preemptively stops unsolvable instances (risking premature stoppage). Given a target risk and a validation set, we use distribution-free risk control to optimally specify these stopping mechanisms. Empirical results across diverse reasoning tasks and models demonstrate the effectiveness of our risk control approach, demonstrating computational efficiency gains from the lower threshold and ensemble stopping mechanisms, all while adhering to the user-specified risk target. Code is available at https://github.com/xidulu/reasoning_risk_control/.
- † Johns Hopkins University
- * Equal contribution
Related readings and updates.
LaDiR: Latent Diffusion Enhances LLMs for Text Reasoning
April 28, 2026research area Speech and Natural Language Processingconference ICLR
Large Language Models (LLMs) demonstrate their reasoning ability through chain-of-thought (CoT) generation. However, LLM’s autoregressive decoding may limit the ability to revisit and refine earlier tokens in a holistic manner, which can also lead to inefficient exploration for diverse solutions. In this paper, we propose LaDiR (Latent Diffusion Reasoner), a novel reasoning framework that unifies the expressiveness of continuous latent…
Recent advances in deep reinforcement learning have demonstrated the capability of learning complex control policies from many types of environments. When learning policies for safety-critical applications, it is essential to be sensitive to risks and avoid catastrophic events. Towards this goal, we propose an actor-critic framework that models the uncertainty of the future and simultaneously learns a policy based on that uncertainty model…
Discover opportunities in Machine Learning.
Our research in machine learning breaks new ground every day.