cs.CL提交于 2026-10-27已译

MANTIS：用于实例级压力测试与最坏情况训练的假新闻检测多轴新闻变换

MANTIS: Multi-Axis News Transformations for Instance-Wise Stress Testing and Worst-Case Training in Fake News Detection

Dexin Liu · Fan Yang · Zimu Wang · Xunfei Zhu · Bolin Ren · Bingyan Zhang · et al.

中文摘要

社交网络中的自动假新闻检测至关重要，因为假新闻可被轻易改写以迷惑检测器。现有方法通常依赖表层写作线索或单风格文本重写，面对多样化外观修改时鲁棒性较差。构建了多轴呈现攻击基准 MANTIS-Bench，将原始新闻与写作风格、情感表达、传播框架等维度的改写版本配对。进一步提出实例自适应对抗框架 MANTIS，可针对每篇新闻动态开展最坏情况训练，且不增加推理时开销。该方法将假新闻防御由启发式格式平滑转化为面向目标且事实不变的真实性评估。MANTIS-Bench 实验显示，在 GossipCop 上，MANTIS 相较 SheepDog 将 Union Attack Success Rate（UASR）降低 3.55 个百分点；其 Macro-F1 持续优于既有单轴防御方法，同时保持具有竞争力的干净准确率，表明多轴与实例自适应最坏情况训练有效。

关键要点

01现有假新闻检测依赖表层写作线索或单风格重写，难以抵御多样化外观修改。
02构建 MANTIS-Bench，从写作风格、情感表达和传播框架三个维度生成改写视图。
03提出实例自适应对抗框架 MANTIS，为每篇新闻动态进行最坏情况训练。
04MANTIS 在 GossipCop 上较 SheepDog 将 UASR 降低 3.55 个百分点。
05MANTIS 的 Macro-F1 优于单轴防御方法，并保持有竞争力的干净准确率。

解读

尚无解读。

原始英文摘要

arXiv:2610.35475v1 Announce Type: new Abstract: Automated fake news detection is vital on social media, where fake news is easily rewritten to fool detectors. However, existing approaches typically rely on superficial writing cues or single-style text rewrites, resulting in poor robustness under diverse cosmetic modifications. We construct MANTIS-Bench, a multi-axis presentation-attack benchmark that pairs original news with rewritten views across writing style, emotional expression, and dissemination framing. We further propose MANTIS, an instance-adaptive adversarial framework enabling dynamic worst-case training tailored to each news article with zero inference-time overhead. This reframes fake news defense from heuristic format smoothing into targeted, fact-invariant credibility assessment. Experiments on MANTIS-Bench show that MANTIS reduces the Union Attack Success Rate (UASR) by 3.55 percentage points relative to SheepDog on GossipCop and consistently outperforms prior single-axis defenses on Macro-F1 while maintaining competitive clean accuracy, demonstrating the effectiveness of multi-axis and instance-adaptive worst-case training.