Is Crowdsourcing Breaking Your Bank? Cost-Effective Fine-Tuning of Pre-trained Language Models with Proximal Policy Optimization

arXiv:2402.18284v1 Announce Type: new
Abstract: Wide usage of ChatGPT has highlighted the potential of reinforcement learning from human feedback. However, its training pipeline relies on manual ranking, a resource-intensive process. To reduce labor costs, we propose a self-supervised text ranking approach for applying Proximal-Policy-Optimization to fine-tune language models while eliminating the need for human annotators. Our method begins with probabilistic sampling to encourage a language model to generate diverse responses for each input. We then employ TextRank and ISODATA algorithms …

abstract arxiv bank breaking chatgpt cost costs crowdsourcing cs.ai cs.cl feedback fine-tuning human human feedback labor language language models optimization pipeline policy process ranking reduce reinforcement reinforcement learning text text ranking training training pipeline type usage