arXiv cs.CY· Alexander Loth, Martin Kappes, Marc-Oliver Pahl·· 11 小时前AI 评分31
JudgeGPT 与 RogueGPT 框架因果分析：人类为何难辨基础模型幻觉与虚假信息
Eroding the Truth-Default: A Causal Analysis of Human Susceptibility to Foundation Model Hallucinations and Disinformation in the Wild
AI 导读
研究提出 JudgeGPT 与 RogueGPT 双轴框架，将"真实性"与"归因"解耦，基于 5 个基础模型（含 GPT-4 和 Llama-2）的 918 次评估，用结构因果模型分析人类识别准确率。
正文 · 原文
Abstract:As foundation models (FMs) approach human-level fluency, distinguishing synthetic from organic content has become a key challenge for Trustworthy Web Intelligence. This paper presents JudgeGPT and RogueGPT, a dual-axis framework that decouples "authenticity" from "attribution" to investigate the mechanisms of human susceptibility. Analyzing 918 evaluations across five FMs (including GPT-4 and Llama-2), we employ Structural Causal Models (SCMs) as a principal framework for formulating testable causal hypotheses about detection accuracy. Contrary to partisan narratives, we find that political orientation shows a negligible association with detection performance ($r=-0.10$). Instead, "fake news familiarity" emerges as a candidate mediator ($r=0.35$), suggesting that exposure may function as adversarial training for human discriminators. We identify a "fluency trap" where GPT-4 outputs (HumanMachineScore: 0.20) bypass Source Monitoring mechanisms, rendering them indistinguishable from human text. These findings suggest that "pre-bunking" interventions should target cognitive source monitoring rather than demographic segmentation to ensure trustworthy information ecosystems.
| Comments: | 8 pages, 4 figures, plus supplementary appendix. Published in Companion Proceedings of the ACM Web Conference 2026 (WWW Companion '26), pp. 1125-1132 |
| Subjects: | Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC) |
| ACM classes: | I.2.7; H.1.2; H.5.2 |
| Cite as: | arXiv:2601.22871 [cs.CY] |
| (or arXiv:2601.22871v3 [cs.CY] for this version) | |
| https://doi.org/10.48550/arXiv.2601.22871 arXiv-issued DOI via DataCite |
|
| Journal reference: | Companion Proceedings of the ACM Web Conference 2026 (WWW Companion '26), pp. 1125-1132 |
| Related DOI: | https://doi.org/10.1145/3774905.3795832
DOI(s) linking to related resources |
Submission history
From: Alexander Loth [view email]
[v1]
Fri, 30 Jan 2026 11:49:58 UTC (1,074 KB)
[v2]
Thu, 12 Feb 2026 08:07:44 UTC (1,074 KB)
[v3]
Thu, 29 Oct 2026 19:35:57 UTC (1,075 KB)
来源：arXiv cs.CY · arxiv.org