AtPatch: Debugging Transformers via Hot-Fixing Over-Attention

TL;DR

AtPatch is a dynamic hot-fix method that detects and corrects anomalous attention patterns in transformer models during inference, effectively mitigating backdoor attacks and unfairness without retraining.

Contribution

It introduces a novel attention redistribution technique that operates on-the-fly during inference, improving mitigation of malicious and biased attention patterns in transformer models.

Findings

More effective at mitigating backdoor attacks.

Better preserves original model functionality.

Works without retraining or modifying model parameters.

Abstract

Transformer-based deep neural networks (DNNs) affected by backdoor attacks and unfairness typically exhibit anomalous attention patterns, leading to over-attend to backdoor triggers or protected attributes. Existing neuron-editing mitigation strategies often struggle to handle such situation and most of them lack flexibility and tend to distort feature representations. Motivated by such over-attention phenomenon and software engineering paradigms such as delta debugging and hot patching, we propose AtPatch, a hot-fix method that dynamically redistributes attention maps during model inference. Specifically, for a given input, AtPatch first extracts the attention map from the model's inference process. Then, it uses a pre-trained detector to identify anomalous columns and replace them with unified benign attention. Then, AtPatch rescales other columns to mitigate the impact of over-attention. Finally, AtPatch returns the redistributed attention map to the model for continued inference. Notably, if the detector does not report any anomalous columns, AtPatch directly returns the original attention map to the model. Unlike existing techniques, AtPatch selectively redistributes the attention map, making it better at preserving the model's original functionality. Furthermore, AtPatch's on-the-fly nature allows it to work without modifying model parameters or retraining, making it better suited for deployed models. We conducted extensive experiments to validate AtPatch. Experimental results show that, compared to existing methods, AtPatch can more effectively mitigate backdoor attacks and unfairness while better preserving the model's original functionality.