# Do Language Models Learn Semantics of Code? A Case Study in Vulnerability Detection

arXiv: https://arxiv.org/abs/2311.04109 · HTML: https://mlsanity.com/paper/2311.04109/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 42 references that are in the map.
- First posted on arXiv 7 Nov 2023.

## Builds on (most cited first, 18 of 18)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Learning Important Features Through Propagating Activation Differences](https://mlsanity.com/paper/1704.02685/) (2017)
- [Towards A Rigorous Science of Interpretable Machine Learning](https://mlsanity.com/paper/1702.08608/) (2017)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation](https://mlsanity.com/paper/2109.00859/) (2021)
- [CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation](https://mlsanity.com/paper/2102.04664/) (2021)
- [Attention is not not Explanation](https://mlsanity.com/paper/1908.04626/) (2019)
- [Captum: A unified and generic model interpretability library for PyTorch](https://mlsanity.com/paper/2009.07896/) (2020)
- [UniXcoder: Unified Cross-Modal Pre-training for Code Representation](https://mlsanity.com/paper/2203.03850/) (2022)
- [Not Just a Black Box: Learning Important Features Through Propagating Activation Differences](https://mlsanity.com/paper/1605.01713/) (2016)
- [BERTology Meets Biology: Interpreting Attention in Protein Language Models](https://mlsanity.com/paper/2006.15222/) (2020)
- [An Empirical Study of Deep Learning Models for Vulnerability Detection](https://mlsanity.com/paper/2212.08109/) (2022)
- [CODIT: Code Editing with Tree-Based Neural Models](https://mlsanity.com/paper/1810.00314/) (2018)
- [On Multi-Modal Learning of Editing Source Code](https://mlsanity.com/paper/2108.06645/) (2021)
- [An Empirical Comparison of Pre-Trained Models of Source Code](https://mlsanity.com/paper/2302.04026/) (2023)
- [Follow-up Attention: An Empirical Study of Developer and Neural Model Code Exploration](https://mlsanity.com/paper/2210.05506/) (2022)

## Built on by (most cited first, 4 of 4)

- [Vulnerability Detection with Code Language Models: How Far Are We?](https://mlsanity.com/paper/2403.18624/) (2024)
- [A Systematic Study of Code Obfuscation Against LLM-based Vulnerability Detection](https://mlsanity.com/paper/2512.16538/) (2025)
- [LPASS: Linear Probes as Stepping Stones for vulnerability detection using compressed LLMs](https://mlsanity.com/paper/2505.24451/) (2025)
- [Probing the Prefill: Detecting Code Vulnerabilities via Latent Activations](https://mlsanity.com/paper/2608.16970/) (2026)

## Abstract (arXiv)

Recently, pretrained language models have shown state-of-the-art performance on the vulnerability detection task. These models are pretrained on a large corpus of source code, then fine-tuned on a smaller supervised vulnerability dataset. Due to the different training objectives and the performance of the models, it is interesting to consider whether the models have learned the semantics of code relevant to vulnerability detection, namely bug semantics, and if so, how the alignment to bug semantics relates to model performance. In this paper, we analyze the models using three distinct methods: interpretability tools, attention analysis, and interaction matrix analysis. We compare the models' influential feature sets with the bug semantic features which define the causes of bugs, including buggy paths and Potentially Vulnerable Statements (PVS). We find that (1) better-performing models also aligned better with PVS, (2) the models failed to align strongly to PVS, and (3) the models failed to align at all to buggy paths. Based on our analysis, we developed two annotation methods which highlight the bug semantics inside the model's inputs. We evaluated our approach on four distinct transformer models and four vulnerability datasets and found that our annotations improved the models' performance in the majority of settings - 11 out of 16, with up to 9.57 points improvement in F1 score compared to conventional fine-tuning. We further found that with our annotations, the models aligned up to 232% better to potentially vulnerable statements. Our findings indicate that it is helpful to provide the model with information of the bug semantics, that the model can attend to it, and motivate future work in learning more complex path-based bug semantics. Our code and data are available at https://figshare.com/s/4a16a528d6874aad51a0.