[论文] Reflective Context Learning: Studying the Optimization Primitives of C...

论文概要

研究领域: ML 作者: Nikita Vassilyev, William Berrios, Ruowang Zhang 等 发布时间: 2026-04-03 arXiv: 2604.03189

中文摘要

通用能力智能体必须以跨任务和环境泛化的方式从经验中学习。学习的基本问题，包括信用分配、过拟合、遗忘、局部最优和高方差学习信号，无论学习对象位于参数空间还是上下文空间都持续存在。我们提出反射上下文学习，一个智能体通过重复交互、对行为和失败模式的反思以及上下文的迭代更新来学习的统一框架。

原文摘要

Generally capable agents must learn from experience in ways that generalize across tasks and environments. The fundamental problems of learning, including credit assignment, overfitting, forgetting, local optima, and high-variance learning signals, persist whether the learned object lies in parameter space or context space. While these challenges are well understood in classical machine learning optimization, they remain underexplored in context space, leading current methods to be fragmented and ad hoc. We present Reflective Context Learning (RCL), a unified framework for agents that learn through repeated interaction, reflection on behavior and failure modes, and iterative updates to context. In RCL, reflection converts trajectories and current context into a directional update signal an...

自动采集于 2026-04-06

#论文 #arXiv #ML #小凯