Study of Distractors in Neural Models of Code

Abstract

Finding important features that contribute to the prediction of neural models is an active area of research in explainable AI. Neural models are opaque and finding such features sheds light on a better understanding of their predictions. In contrast, in this work, we present an inverse perspective of distractor features: features that cast doubt about the prediction by affecting the model's confidence in its prediction. Understanding distractors provide a complementary view of the features' relevance in the predictions of neural models. In this paper, we apply a reduction-based technique to find distractors and provide our preliminary results of their impacts and types. Our experiments across various tasks, models, and datasets of code reveal that the removal of tokens can have a significant impact on the confidence of models in their predictions and the categories of tokens can also play a vital role in the model's confidence. Our study aims to enhance the transparency of models by emphasizing those tokens that significantly influence the confidence of the models.

arXiv commentThe 1st International Workshop on Interpretability and Robustness in Neural Software Engineering, Co-located with ICSE (InteNSE'23)

References in corpus (14)

• Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications
• CodeSearchNet Challenge: Evaluating the State of Semantic Code Search
• CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation
• Embedding Java Classes with code2vec: Improvements from Variable Obfuscation
• Neural Program Repair by Jointly Learning to Localize and Repair
• Diet Code Is Healthy: Simplifying Programs for Pre-trained Models of Code
• Understanding Neural Code Intelligence Through Program Simplification
• Towards security defect prediction with AI
• Memorization and Generalization in Neural Code Intelligence Models
• COSET: A Benchmark for Evaluating Neural Program Embeddings
• Syntax-Guided Program Reduction for Understanding Neural Code Intelligence Models
• Testing Neural Program Analyzers
• Code2Snapshot: Using Code Snapshots for Learning Representations of Source Code
• WheaCha: A Method for Explaining the Predictions of Models of Code

Cited by in corpus (1)

• Your Code Secret Belongs to Me: Neural Code Completion Tools Can Memorize Hard-Coded Credentials