AI 中文总结
本文提出RoE-FND框架，结合LLM与自反经验学习构建经验库，推理时检索经验裁决对立推演，在5个FND基准上优于基线且跨数据集泛化性强。
AI 中文摘要
社交网络中虚假内容的泛滥催生了对稳健假新闻检测（FND）系统的需求。现有方案要么在标注数据上训练检测器，要么利用大语言模型（LLMs）的推理能力，但当前方法要么泛化性受限，要么易过度依赖有说服力却存在缺陷的推理依据，缺乏系统经验与机制来揭露细微的推理错误。本文提出RoE-FND（基于经验推理的假新闻检测框架），这一基于LLM的框架将自反式经验构建与通过检索经验进行的审议相结合以实现FND。RoE-FND通过自反学习构建经验库：以真实标签作为后验监督，对比无约束分析与标签条件分析，再将二者的关键分歧总结为可复用的推理准则。推理阶段，RoE-FND通过作为后验的翻转伪标签生成两种对立推演，检索最相关的经验以解决二者的关键分歧，并裁定支撑更充分的推理依据作为最终预测。在五个流行基准（仅文本数据集CHEF、Snopes、PolitiFact，以及多媒体数据集FakeTT、FakeSV）上的实验表明，RoE-FND在不针对数据集分布优化LLM参数的情况下，性能优于强基线，且展现出强大的跨数据集泛化能力。
英文摘要
The proliferation of deceptive content in social networks necessitates robust Fake News Detection (FND) systems. Existing pipelines either train detectors on labeled data or leverage Large Language Models (LLMs) for their reasoning ability. However, current approaches remain either limited in generalizability or prone to over-commitment to persuasive yet flawed rationales, lacking systematic experience and mechanisms to expose subtle reasoning errors. We propose \textbf{RoE-FND} (\textbf{\underline{R}}eason \textbf{\underline{o}}n \textbf{\underline{E}}xperiences FND), an LLM-based framework that combines self-reflective experience building with deliberation through retrieved experiences for FND. RoE-FND builds an experience bank via reflective learning that compares an unconstrained analysis with a label-conditioned analysis using the ground-truth label as posterior supervision, then summarizes their critical divergence into reusable reasoning guidelines. During inference, RoE-FND generates two opposing deductions via a flipped pseudo-label provided as posterior, retrieves the most relevant experiences for resolving their key disagreement, and adjudicates the better-supported rationale as the final prediction. Experiments across five popular benchmarks, including text-only datasets, i.e., CHEF, Snopes, PolitiFact, and multimedia datasets, i.e., FakeTT, FakeSV, demonstrate that RoE-FND outperforms strong baselines without optimizing LLM parameters on dataset distributions, while exhibiting strong cross-dataset generalization.
