Textual backdoor attacks pose significant security threats. Current detection
approaches, typically relying on intermediate feature representation or
reconstructing potential triggers, are task-specific and less effective beyond
sentence classification, struggling with tasks like question answering and
named entity recognition. We introduce TABDet (Task-Agnostic Backdoor
Detector), a pioneering task-agnostic method for backdoor detection. TABDet
leverages final layer logits combined with an efficient pooling technique,
enabling unified logit representation across three prominent NLP tasks. TABDet
can jointly learn from diverse task-specific models, demonstrating superior
detection efficacy over traditional task-specific methods.

文本后门攻击是重要的安全威胁。当前的检测方法通常依赖于中间特征表示或重构潜在触发器，任务特异性和在问题回答和命名实体识别等任务中的效果较差。我们介绍了 TABDet（任务不可知后门检测器），一种用于后门检测的开创性的任务不可知方法。TABDet 利用最后一层的 logits 结合高效的池化技术，实现了统一的 logit 表示，适用于三个主要的自然语言处理任务。TABDet 可以从不同的任务特定模型中联合学习，展示出优于传统任务特定方法的检测效果。

无任务依赖的检测器用于插入式背门攻击

Task-Agnostic Detector for Insertion-Based Backdoor Attacks

This paper aims to analyze knowledge consistency between pre-trained deep
neural networks. We propose a generic definition for knowledge consistency
between neural networks at different fuzziness levels. A task-agnostic method
is designed to disentangle feature components, which represent the consistent
knowledge, from raw intermediate-layer features of each neural network. As a
generic tool, our method can be broadly used for different applications. In
preliminary experiments, we have used knowledge consistency as a tool to
diagnose representations of neural networks. Knowledge consistency provides new
insights to explain the success of existing deep-learning techniques, such as
knowledge distillation and network compression. More crucially, knowledge
consistency can also be used to refine pre-trained networks and boost
performance.

本文旨在分析预训练深度神经网络之间的知识一致性，并提出了一个泛化定义。方式是设计了一种任务不可知的方法来解开每个神经网络的原始中间层特征和表示一致知识的特征组成部分。这项工作可以用于不同应用，例如调整预训练网络以提高性能并解释基于知识蒸馏和网络压缩等深度学习技术成功的原因。