arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.17271cs.AI
收藏

基于标签引导的思维链优化的宣传检测推理

Reasoning for Propaganda Detection via Label-Guided CoT Optimisation

Indraprastha Institute of Information Technology Delhi(因德拉普拉斯塔信息技术学院德里校区)

Zeba Afroz, Harsh Vardhan, Md. Shad Akhtar

AI总结:

本文针对宣传检测系统缺乏可解释性的问题,提出ReCoT框架,通过群体优化与性能重采样捕捉说服性技巧的可定义线索,在INDI-PROPspan和PTC数据集上均取得优于基线的结果,且保持可解释性。

摘要:

宣传检测系统通常作为黑箱运行,缺乏可解释性。为解决该问题,本文假设说服性技巧针对其社会政治环境具有可定义特征。本文提出ReCoT,一种推理条件检测框架,该框架通过群体层面的优化和性能引导的重采样捕捉说服性技巧的可定义线索,无需额外的解释标注。在INDI-PROPspan数据集上,ReCoT的宏F1_LCS为0.364,优于基线模型,达到Avg F1_LCS的SFT上限的88%;在PTC数据集上,ReCoT达到Avg F1_LCS的SFT上限的85%,大幅缩小了差距,同时保持可解释性。对ReCoT的推理池收敛性和更新模式的分析表明,其能有效优化推理思维链(CoT)。

英文摘要:

Propaganda detection systems usually operate as black boxes, lacking explanations. To address this, we hypothesise that persuasive techniques contain definitional characteristics with respect to their social-political setting. We introduce ReCoT, a reasoning-conditioned detection framework that captures definitional cues of persuasive techniques via population-level refinement and performance-guided resampling, without requiring additional explanation annotations. On INDI-PROPspan, ReCoT achieves a macro F1_LCS of 0.364, surpassing baselines and reaching 88% of the SFT upper bound on Avg F1_LCS. Similarly, on PTC, ReCoT reaches 85% of the SFT upper bound on Avg F1_LCS, closing much of this gap while remaining interpretable. Importantly, the analysis of the reasoning pool's convergence and update pattern for ReCoT shows effective refinement of reasoning CoTs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑