Text-to-Code Generation with Modality-relative Pre-training
Christopoulou, Fenia, Zhang, Guchun, Lampouras, Gerasimos
–arXiv.org Artificial Intelligence
Large pre-trained language models have recently been expanded and applied to programming language tasks with great success, often through further pre-training of a strictly-natural language model--where training sequences typically contain both natural and (linearised) programming language. Such approaches effectively map both modalities of the sequence into the same embedding space. However, programming language keywords (e.g. "while") often have very strictly defined semantics. As such, transfer learning from their natural language usage may not necessarily be beneficial to their code application and vise versa. Assuming an already pre-trained language model, in this work we investigate how sequence tokens can be adapted and represented differently, depending on which modality they belong to, and to the ultimate benefit of the downstream task. We experiment with separating embedding spaces between modalities during further model pre-training with modality-relative training objectives. We focus on text-to-code generation and observe consistent improvements across two backbone models and two test sets, measuring pass@$k$ and a novel incremental variation.
arXiv.org Artificial Intelligence
Feb-12-2024
- Country:
- North America
- Dominican Republic (0.04)
- United States
- New York > New York County
- New York City (0.04)
- Minnesota > Hennepin County
- Minneapolis (0.14)
- California > San Diego County
- San Diego (0.04)
- New York > New York County
- Europe
- United Kingdom > England
- Greater London > London (0.04)
- Ireland > Leinster
- County Dublin > Dublin (0.04)
- Belgium > Brussels-Capital Region
- Brussels (0.04)
- United Kingdom > England
- North America
- Genre:
- Research Report (0.50)
- Industry:
- Health & Medicine (0.93)
- Technology: