Summary of Weak-to-strong Reasoning, by Yuqing Yang et al.
Weak-to-Strong Reasoning
by Yuqing Yang, Yan Ma, Pengfei Liu
First submitted to arxiv on: 18 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a progressive learning framework that enables large language models (LLMs) to autonomously refine their training data, without requiring input from either a more advanced model or human-annotated data. The method starts with supervised fine-tuning on a selective small but high-quality dataset, followed by preference optimization on contrastive samples identified by the strong model itself. Experimental results show that this approach significantly enhances the reasoning capabilities of Llama2-70b using three separate weak models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper introduces a new way for large language models to learn and improve without needing help from more advanced models or humans. The method uses a combination of supervised learning and optimization techniques to refine the model’s training data. This helps the model learn better and make more accurate decisions. The results show that this approach can be very effective in improving the performance of Llama2-70b. |
Keywords
» Artificial intelligence » Fine tuning » Optimization » Supervised
