Efficient Code Embeddings from Code Generation Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912558812758016 |
|---|---|
| author | Kryvosheieva, Daria Sturua, Saba Günther, Michael Martens, Scott Xiao, Han |
| author_facet | Kryvosheieva, Daria Sturua, Saba Günther, Michael Martens, Scott Xiao, Han |
| contents | jina-code-embeddings is a novel code embedding model suite designed to retrieve code from natural language queries, perform technical question-answering, and identify semantically similar code snippets across programming languages. It makes innovative use of an autoregressive backbone pre-trained on both text and code, generating embeddings via last-token pooling. We outline the training recipe and demonstrate state-of-the-art performance despite the relatively small size of the models, validating this approach to code embedding model construction. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_21290 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Efficient Code Embeddings from Code Generation Models Kryvosheieva, Daria Sturua, Saba Günther, Michael Martens, Scott Xiao, Han Computation and Language Artificial Intelligence Information Retrieval 68T50 I.2.7 jina-code-embeddings is a novel code embedding model suite designed to retrieve code from natural language queries, perform technical question-answering, and identify semantically similar code snippets across programming languages. It makes innovative use of an autoregressive backbone pre-trained on both text and code, generating embeddings via last-token pooling. We outline the training recipe and demonstrate state-of-the-art performance despite the relatively small size of the models, validating this approach to code embedding model construction. |
| title | Efficient Code Embeddings from Code Generation Models |
| topic | Computation and Language Artificial Intelligence Information Retrieval 68T50 I.2.7 |
| url | https://arxiv.org/abs/2508.21290 |