用大模型自动发明逻辑编程的新谓词，让机器自己找规律。
ADVENT: LLM-Driven Automatic Predicate Invention for ILP
- 大模型生成候选谓词，程序验证并迭代优化
- 在扑克牌任务中成功率提升至80%，比纯逻辑编程高58%
- 发明的谓词可跨任务复用，结果可读性强
谓词发明（PI）是归纳逻辑编程（ILP）中的关键瓶颈。现有方法依赖领域知识，生成的谓词语义不透明，难以适应新领域或跨任务复用。我们提出ADVENT，一种基于大语言模型（LLM）的ILP谓词发明机制。该机制结合大模型的溯因生成与Prolog的演绎验证，形成迭代循环，由具体执行结果引导大模型优化候选谓词。该方法利用大模型发现结构化关系数据中的隐含模式，生成命名清晰、定义明确的辅助谓词。发明的谓词与学习规则存入知识池，支持跨任务复用。在七个大模型上对九个扑克手牌概念的实验表明，大模型驱动的谓词发明达到58%的成功率，而纯ILP完全失败；经形式化验证后成功率提升至80%，知识池带来最高+31个百分点的性能增益，且生成规则具有人类可读性。结果表明，ADVENT为自动化谓词发明和实现ILP中跨任务知识复用提供了有前景的方向。
原文摘要 · Abstract (English)
Predicate invention (PI), the creation of new predicates to extend the hypothesis space, remains a critical bottleneck in Inductive Logic Programming (ILP). Existing methods rely on domain expertise and produce semantically opaque predicates, hindering adaptation to unfamiliar domains and cross-task reuse. We present ADVENT, an LLM-driven PI mechanism for ILP. ADVENT pairs LLM abductive generation with Prolog deductive verification, forming an iterative loop in which concrete execution results guide the LLM to refine candidate predicates. The mechanism leverages Large Language Models to identify implicit patterns in structured relational data and invent auxiliary predicates with meaningful names and definitions. Invented predicates and learned rules accumulate in a knowledge pool for cross-task reuse. Experiments on nine poker-hand concepts across seven LLMs show that LLM-driven PI achieves 58% success rate where ILP alone fails entirely, formal verification raises this to 80%, and the knowledge pool yields gains up to +31 percentage points, while producing human-interpretable rules. These results suggest that ADVENT offers a promising direction for automating predicate invention and enabling cross-task knowledge reuse in ILP.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。