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

arXiv: https://arxiv.org/abs/2402.18284 · HTML: https://mlsanity.com/paper/2402.18284/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 27 of its 52 references that are in the map.
- First posted on arXiv 28 Feb 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 27 of 27)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [GLM-130B: An Open Bilingual Pre-trained Model](https://mlsanity.com/paper/2210.02414/) (2022)
- [DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset](https://mlsanity.com/paper/1710.03957/) (2017)
- [ERNIE: Enhanced Language Representation with Informative Entities](https://mlsanity.com/paper/1905.07129/) (2019)
- [Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Models](https://mlsanity.com/paper/2108.08877/) (2021)
- [AdapterHub: A Framework for Adapting Transformers](https://mlsanity.com/paper/2007.07779/) (2020)
- [SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient](https://mlsanity.com/paper/1609.05473/) (2016)
- [How Language Model Hallucinations Can Snowball](https://mlsanity.com/paper/2305.13534/) (2023)
- [Making Large Language Models Better Reasoners with Step-Aware Verifier](https://mlsanity.com/paper/2206.02336/) (2022)
- [Exploring Versatile Generative Language Model Via Parameter-Efficient Transfer Learning](https://mlsanity.com/paper/2004.03829/) (2020)
- [Pre-trained Language Model Representations for Language Generation](https://mlsanity.com/paper/1903.09722/) (2019)
- [User Intent Recognition and Satisfaction with Large Language Models: A User Study with ChatGPT](https://mlsanity.com/paper/2402.02136/) (2024)
- [An Unsupervised Semantic Sentence Ranking Scheme for Text Documents](https://mlsanity.com/paper/2005.02158/) (2020)

## Built on by (most cited first, 2 of 2)

- [Detecting Instruction Fine-tuning Attacks using Influence Function](https://mlsanity.com/paper/2504.09026/) (2025)
- [Improving Targeted Molecule Generation through Language Model Fine-Tuning Via Reinforcement Learning](https://mlsanity.com/paper/2405.06836/) (2024)

## Abstract (arXiv)

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 to rank and cluster these responses based on their semantics. Subsequently, we construct a reward model to learn the rank and optimize our generative policy. Our experimental results, conducted using two language models on three tasks, demonstrate that the models trained by our method considerably outperform baselines regarding BLEU, GLEU, and METEOR scores. Furthermore, our manual evaluation shows that our ranking results exhibit a remarkably high consistency with that of humans. This research significantly reduces training costs of proximal policy-guided models and demonstrates the potential for self-correction of language models.