arXiv cs.AI· Giacomo Magnifico, Eduard Barbu·· 2 天前AI 评分24
ReX-GG：面向关系抽取与图谱生成的 LLM 集成流水线
ReX-GG: a LLM Ensemble Pipeline for Relation-extraction and Graph Generation
AI 导读
ReX-GG 是一个可定制的轻量级 LLM 协同集成流水线，通过 JSON 结构化输出与匿名同行评审排序来抑制模型幻觉和单点故障。该流水线在英文科普文章的关系抽取任务上演示，覆盖强因果、弱因果、对比和组合四类本体关系，最终输出经语义节点规范化的彩色 HTML 因果图谱。匿名用户研究显示其相对人工标注金标准的平均感知准确率为 0.778，框架以开放许可证发布，支持 API 与本地 LLM 部署。
正文 · 原文
Abstract:Current LLM ensemble frameworks focus on multi-step setups with additional modules for answer ranking, often opting for token and span analysis rather than structured outputs, leading to heavyweight architectures with potential fail states along the pipeline. Faster, lighter solutions are more vulnerable to hallucination propagation and can lack output control in more complex pipelines. This paper proposes a customisable, lightweight ensemble workflow of coordinated Large Language Models that leverages JSON-structured outputs and anonymous peer-review ranking to mitigate hallucinatory outputs and single-model failure points. The pipeline is demonstrated on a relation extraction task applied to popular science articles in English, targeting four ontologically-grounded relation types (strong causation, weak causation, contrastive, and compositional), with semantic node canonicalisation and interactive, colour-coded HTML causal graphs as the final output. Performance is evaluated through an anonymous user study, achieving an average perceived accuracy of 0.778 against a human-annotated gold standard. The modular architecture supports flexible deployment across both API-based and in-house LLM setups, and the full framework is released under an open license to foster reproducibility and collaborative research.
| Comments: | 9 pages, 4 figures, accepted at KG&LLM @ LREC 2026 |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.41901 [cs.AI] |
| (or arXiv:2610.41901v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41901 arXiv-issued DOI via DataCite (pending registration) |
|
| Journal reference: | Proceedings of the Knowledge Graphs and Large Language Models Workshop (KG-LLM) @ LREC26, Palma, Mallorca, Spain, 2026, pp. 93-101 |
| Related DOI: | https://doi.org/10.63317/2qkh8idwebsv
DOI(s) linking to related resources |
Submission history
From: Giacomo Magnifico [view email]
[v1]
Fri, 30 Oct 2026 15:39:02 UTC (386 KB)
来源：arXiv cs.AI · arxiv.org