Demo-Craft: Using In-Context Learning to Improve Code Generation in Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917964817629184 |
|---|---|
| author | Kapu, Nirmal Joshua Sreejith, Mihit |
| author_facet | Kapu, Nirmal Joshua Sreejith, Mihit |
| contents | Generating executable code from natural language instructions using Large Language Models (LLMs) poses challenges such as semantic ambiguity and understanding taskspecific contexts. To address these issues, we propose a system called DemoCraft, which enhances code generation by leveraging in-context learning and demonstration selection, combined with latent concept learning. Latent concept learning introduces additional concept tokens, which are trainable embeddings that capture task-specific knowledge. We then test our system on two major datasets: MBPP and Humaneval. Our experimental results demonstrate that the proposed system achieves an approximate 2x increase in the pass@k metric compared to baseline models. Furthermore, we introduce two novel evaluation metrics: correctness@k and similarity@k. Our empirical studies indicate that our system attains nearly a 3x improvement in these metrics as well. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_00865 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Demo-Craft: Using In-Context Learning to Improve Code Generation in Large Language Models Kapu, Nirmal Joshua Sreejith, Mihit Software Engineering Artificial Intelligence Computation and Language Generating executable code from natural language instructions using Large Language Models (LLMs) poses challenges such as semantic ambiguity and understanding taskspecific contexts. To address these issues, we propose a system called DemoCraft, which enhances code generation by leveraging in-context learning and demonstration selection, combined with latent concept learning. Latent concept learning introduces additional concept tokens, which are trainable embeddings that capture task-specific knowledge. We then test our system on two major datasets: MBPP and Humaneval. Our experimental results demonstrate that the proposed system achieves an approximate 2x increase in the pass@k metric compared to baseline models. Furthermore, we introduce two novel evaluation metrics: correctness@k and similarity@k. Our empirical studies indicate that our system attains nearly a 3x improvement in these metrics as well. |
| title | Demo-Craft: Using In-Context Learning to Improve Code Generation in Large Language Models |
| topic | Software Engineering Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2411.00865 |