发表机构
University of Illinois Urbana-Champaign; Yale University; The Pennsylvania State University（伊利诺伊大学厄巴纳-香槟分校; 耶鲁大学; 宾夕法尼亚州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
PAPER2LLM++提出从研究论文中持续自我进化LLM的框架，通过提取证据、测试局限并采用尝试-评估-提交机制整合更新，使模型逐步学习新发现而保留旧成果。
AI 中文摘要
关于大语言模型（LLM）的研究不断揭示模型的局限性、其成因以及潜在的解决方案。然而，这些人类发现与模型进化在很大程度上是脱节的：大语言模型不会自动从关于其自身失败的新研究中学习。我们提出了PAPER2LLM++，一个从研究论文中实现大语言模型持续自我进化的框架。PAPER2LLM++并不将论文仅仅视为待检索的知识，而是将不断增长的文献作为模型改进的证据和监督流。对于每篇新论文，它提取基于证据的发现，测试所报告的局限性在当前模型中是否仍然存在，并在需要时将发现转化为候选学习信号。一个“尝试-评估-提交”程序仅在改进目标行为且不会大幅遗忘先前改进或削弱通用能力时整合更新。在顺序流式的研究发现的LLM失败案例中，我们展示了模型能够逐步纳入新发现，同时保留早期成果。因此，PAPER2LLM++朝着弥合人类发现与模型进化之间鸿沟迈出了一步，使模型能够不断从关于自身局限性和改进的研究中学习。
英文摘要
Research on LLMs continually uncovers model limitations, their causes, and potential solutions. Yet these human discoveries remain largely disconnected from model evolution: an LLM does not automatically learn from new research about its own failures. We introduce PAPER2LLM++, a framework for continual self-evolution of LLMs from research papers. Rather than treating papers merely as knowledge to retrieve, PAPER2LLM++ uses the growing literature as a stream of evidence and supervision for model improvement. For each incoming paper, it extracts evidence-grounded findings, tests whether the reported limitation persists in the current model, and, when needed, converts the findings into candidate learning signals. A try-evaluate-commit procedure integrates an update only when it improves the targeted behavior without substantially forgetting prior improvements or degrading general capabilities. Across a sequential stream of research-discovered LLM failures, we show that models can progressively incorporate new findings while retaining earlier gains. PAPER2LLM++ thus takes a step toward closing the loop between human discovery and model evolution, enabling models to continually learn from research about their own limitations and improvements.
Comments20 pages, 5 figures, 12 tables
