池化注意力：评估预训练变换器嵌入用于欺骗分类
Pooling Attention: Evaluating Pretrained Transformer Embeddings for Deception Classification
AI总结：
本文通过评估预训练变换器嵌入在欺骗分类中的效果，发现BERT结合逻辑回归在LIAR数据集上表现最佳，证明了上下文自注意力编码的有效性。
摘要：
本文研究了虚假新闻检测作为Transformer表示的下游评估，通过基准测试将仅编码器和仅解码器的预训练模型（BERT、GPT-2、Transformer-XL）作为冻结嵌入器与轻量级分类器配对进行比较。通过受控预处理比较池化与填充以及神经与线性头部，结果表明上下文自注意力编码一致地有效转移。BERT嵌入结合逻辑回归在LIAR数据集分割上优于神经基线，而序列长度和聚合分析揭示了对截断的鲁棒性和简单最大或平均池化的优势。本工作将基于注意力的token编码器定位为稳健、以架构为中心的基础，隔离Transformer的贡献免受分类器复杂性的干扰。
英文摘要：
This paper investigates fake news detection as a downstream evaluation of Transformer representations, benchmarking encoder-only and decoder-only pre-trained models (BERT, GPT-2, Transformer-XL) as frozen embedders paired with lightweight classifiers. Through controlled preprocessing comparing pooling versus padding and neural versus linear heads, results demonstrate that contextual self-attention encodings consistently transfer effectively. BERT embeddings combined with logistic regression outperform neural baselines on LIAR dataset splits, while analyses of sequence length and aggregation reveal robustness to truncation and advantages from simple max or average pooling. This work positions attention-based token encoders as robust, architecture-centric foundations for veracity tasks, isolating Transformer contributions from classifier complexity.
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。
