Applying RLAIF for Code Generation with API-usage in Lightweight LLMs
This addresses the challenge of hallucination in LLMs for code generation tasks, particularly for developers using resource-constrained models, though it is incremental as it adapts an existing RLAIF method to a specific domain.
The paper tackles the problem of improving code generation with API calls in lightweight LLMs by applying Reinforcement Learning from AI Feedback (RLAIF), resulting in a 4.5% improvement in executability rate and enabling a 780M-parameter model to outperform a 7B-parameter baseline by 1.0%.
Reinforcement Learning from AI Feedback (RLAIF) has demonstrated significant potential across various domains, including mitigating harm in LLM outputs, enhancing text summarization, and mathematical reasoning. This paper introduces an RLAIF framework for improving the code generation abilities of lightweight (<1B parameters) LLMs. We specifically focus on code generation tasks that require writing appropriate API calls, which is challenging due to the well-known issue of hallucination in LLMs. Our framework extracts AI feedback from a larger LLM (e.g., GPT-3.5) through a specialized prompting strategy and uses this data to train a reward model towards better alignment from smaller LLMs. We run our experiments on the Gorilla dataset and meticulously assess the quality of the model-generated code across various metrics, including AST, ROUGE, and Code-BLEU, and develop a pipeline to compute its executability rate accurately. Our approach significantly enhances the fine-tuned LLM baseline's performance, achieving a 4.5% improvement in executability rate. Notably, a smaller LLM model (780M parameters) trained with RLAIF surpasses a much larger fine-tuned baseline with 7B parameters, achieving a 1.0% higher code executability rate.