DS@GT ARC at CheckThat! 2026: A Verdict-Anchored Pipeline for Generating Fact-Checking Articles

Computer Science > Computation and Language arXiv:2611.05382 (cs) [Submitted on 30 Jul 2026] Title: DS@GT ARC at CheckThat! 2026: A Verdict-Anchored Pipeline for Generating Fact-Checking Articles Authors: Hoang Thanh Thanh Truong, Shuyu Tian.

计算机科学 > 计算与语言 arXiv:2611.05382 (cs) [提交于 2026 年 7 月 30 日] 标题：DS@GT ARC 在 CheckThat! 2026：一种用于生成事实核查文章的判决锚定流水线 作者：Hoang Thanh Thanh Truong, Shuyu Tian。

Abstract: Misinformation spreads faster than professional fact-checkers can manually verify it and write articles explaining their verdicts. This creates a need to automate the production of fact-checking articles. This paper details work done by the DS@GT ARC team for CLEF 2026 CheckThat! Lab Task 3 Fact-Checking Article Generation.

摘要：虚假信息的传播速度超过了专业事实核查人员手动核实并撰写文章解释其判决的速度。这产生了对自动化生成事实核查文章的需求。本文详细介绍了 DS@GT ARC 团队为 CLEF 2026 CheckThat! 实验室任务 3（事实核查文章生成）所做的工作。

We seek to generate a full fact-checking article from a claim, its original verdict, and a set of scraped source documents. Our team built a four-stage pipeline that cleans noisy evidence, retrieves and clusters relevant sources, generates the article by conditioning Llama-3.1-8B-Instruct on the original fact-checker verdict, and recovers inline citations through deterministic post-processing.

我们旨在根据声明、其原始判决以及一组抓取的源文档来生成完整的事实核查文章。我们的团队构建了一个四阶段流水线，用于清理噪声证据、检索并聚类相关来源、通过将 Llama-3.1-8B-Instruct 置于原始事实核查判决的条件下生成文章，并通过确定性后处理恢复内联引用。

We achieved a mean score of 0.428 and ranked 3rd out of 9 teams for the CLEF 2026 CheckThat! Lab Task 3, compared to a baseline mean score of 0.272. Our code is available on GitHub at this https URL.

我们在 CLEF 2026 CheckThat! 实验室任务 3 中获得了 0.428 的平均分，在 9 个团队中排名第 3，而基准平均分为 0.272。我们的代码可在 GitHub 上通过此 https URL 获取。