发表机构
University of Luxembourg; Carnegie Mellon University（卢森堡大学; 卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出MERITED框架，结合基于实例的学习理论与能量模型，实现基于不确定性的动态计算分配，并开源191M参数推理模型。
AI 中文摘要
元认知涉及对认知过程本身的推理。一个例子是资源分配，我们在开始推理任务之前根据信心决定投入多少时间和精力。当前依赖大型语言模型（LLM）的人工智能（AI）系统在未先做出响应的情况下无法估计其输出的不确定性，也无法动态分配资源以产生输出，这使得此类元认知过程难以实现。最近提出的解决这两个问题的经典Transformer架构的替代方案是能量基础模型（EBM），它允许可解释的不确定性建模和计算资源的动态分配。虽然EBM可以控制这两个过程，但基于不确定性确定计算分配的实际元认知任务并未直接解决。基于实例的学习理论（IBLT）提供了一种从经验中建模类人决策的方法，此前已应用于预测人类元认知推理。在本文中，我们介绍了一个基于实例学习理论和能量动力学的元认知推理框架（MERITED）。该框架以IBLT为基础，允许控制EBM中分配的计算工作量，从而基于不确定性对推理工作量进行元认知控制，同时保持计算效率。这项工作有两个主要贡献：训练并开放共享一个191M参数的推理EBM，以及使用IBL模型实现MERITED框架以进行动态计算分配。
英文摘要
Metacognition involves reasoning about cognitive processes themselves. An example is in resource allocation where we choose how much time and effort to put into a reasoning task before we begin based on our confidence. Current Artificial Intelligence (AI) systems that rely on Large Language Models (LLMs) cannot estimate their uncertainty about an output without first responding, and cannot dynamically allocate resources to producing an output, making this type of metacognitive process difficult. A recently proposed alternative to classic transformer architectures that addresses these two concerns is the Energy Based Model (EBM) which allows for interpretable uncertainty modeling and dynamic allocation of compute resources. While EBMs can allow for control of these two processes, the actual metacognitive task of determining compute allocation based on uncertainty is not directly addressed. Instance-Based Learning Theory (IBLT) provides an approach to modeling human-like decisions from experience that has previously been applied to predicting human metacognitive reasoning. In this paper we introduce a framework for MEtacognitive Reasoning with Instance-based Learning Theory and Energy Dynamics (MERITED). Grounded in IBLT, this framework allows for control of the computational effort allocated in an EBM to allow for metacognitive control over reasoning effort based on uncertainty while remaining computationally efficient. This work has two main contributions, the training and open weight sharing of a 191M parameter reasoning EBM, and an implementation of the MERITED framework for dynamic compute allocation using an IBL model.
