Salesforce’s CodeRL Achieves SOTA Code Generation Results With Strong Zero-Shot Transfer Capabilities | Synced
In the new paper CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning, a Salesforce Research team presents CodeRL, a novel framework for program synthesis tas...
Source: Synced | AI Technology & Industry Review
In the new paper CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning, a Salesforce Research team presents CodeRL, a novel framework for program synthesis tasks that employs pretrained language models (LMs) and deep reinforcement learning (RL) and achieves state-of-the-art performance on the challenging APPS benchmark while also demonstrating impressive zero-shot transfer capabilities.