Towards Better Code Understanding in Decoder-Only Models with Contrastive Learning
This addresses the need for efficient code understanding in AI systems, offering a more computationally feasible alternative to training large encoder-only models, though it is incremental as it builds on existing pre-trained models.
The paper tackles the problem of limited performance of decoder-only models on code understanding tasks like code search and clone detection by proposing CL4D, a contrastive learning framework that adapts these models to learn discriminative representations, achieving competitive or superior results on benchmark datasets.
Recent advances in large-scale code generation models have led to remarkable progress in producing high-quality code. These models are trained in a self-supervised manner on extensive unlabeled code corpora using a decoder-only architecture. However, despite their generative strength, decoder-only models often exhibit limited performance on code understanding tasks such as code search and clone detection, primarily due to their generation-oriented training objectives. While training large encoder-only models from scratch on massive code datasets can improve understanding ability but remains computationally expensive and time-consuming. In this paper, we explore a more efficient alternative by transferring knowledge from pre-trained decoder-only code generation models to code understanding tasks. We investigate how decoder-only architectures can be effectively adapted to learn discriminative and semantically meaningful code representations. To this end, we propose CL4D, a contrastive learning framework tailored to strengthen the representation capabilities of decoder-only models. Extensive experiments on multiple benchmark datasets demonstrate that CL4D achieves competitive or superior performance compared to existing methods on representative code understanding tasks, including code search and clone detection. Further analysis reveals that CL4D substantially improves the semantic alignment of code representations by reducing the distance between semantically similar code snippets. These findings highlight the feasibility of leveraging decoder-only models as a unified backbone for both code generation and understanding.