跳到正文
arXiv cs.AI· Siqi Lai, Haitao Li, Hao Liu·· 56 分钟前AI 评分38

JevRSI 框架提出低成本递归自我改进的决策模型方法

Think Once, Act Fast: Cost-Effective Recursive Self-Improvement for Decision Model Harness

AI 导读

arXiv 论文提出 JevRSI,一个 System-1/System-2 协作框架,通过递归式 harness 改进把 LLM 推理转化为可迁移的决策支持。

正文 · 原文

View PDF HTML (experimental)

Abstract:Fast System-1 models enable responsive actions but struggle with complex or unfamiliar tasks. Large language model (LLM)-based System-2 provides strong reasoning capabilities, yet repeated invocation incurs substantial latency and computational costs. To bridge this gap, we propose JevRSI, a System-1/System-2 collaboration framework that transforms LLM reasoning into transferable decision support through recursive harness improvement. A Jev-style System-1 model enables low-latency action selection, while an asynchronous System-2 LLM develops a modular harness. The harness integrates procedures, state memory, and world models to provide decision contexts spanning the past, present, and future. Utility-driven retrieval selects modules based on both contextual relevance and their utility on previous tasks. Environmental feedback guides targeted revisions and enables recursive self-improvement. Experiments across three datasets show that JevRSI consistently improves decision quality, matching or surpassing LLM agents in several settings. It preserves low-latency execution with favorable performance-cost trade-offs. Moreover, System-2 development interventions decrease as the harness evolves, amortizing System-2 costs through continuous improvement. Our project is available at this https URL.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.22432 [cs.AI]
  (or arXiv:2611.22432v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.22432

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Siqi Lai [view email]
[v1] Tue, 17 Nov 2026 09:06:14 UTC (2,878 KB)

来源:arXiv cs.AI · arxiv.org