Summary of How Does Quantization Affect Multilingual Llms?, by Kelly Marchisio et al.
How Does Quantization Affect Multilingual LLMs?
by Kelly Marchisio, Saurabh Dash, Hongyu Chen, Dennis Aumiller, Ahmet Üstün, Sara Hooker, Sebastian Ruder
First submitted to arxiv on: 3 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
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 This paper explores the effects of quantization on large language models (LLMs) across languages and scales. The authors use automatic benchmarks, human evaluation, and LLM-as-a-Judge to analyze the impact of quantization on multilingual LLMs. They find that quantization can have a significant negative effect on model performance, particularly in non-Latin script languages like Japanese. The results suggest that there is a need for considering multilingual performance as an important evaluation criterion when developing efficient models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study looks at how reducing the precision of language models affects their ability to understand and generate text across different languages. Researchers used special tests, asked humans to evaluate the models, and even had the models judge themselves (LLM-as-a-Judge). They found that making language models smaller can make them worse, especially for languages like Japanese. This is important because we need better ways to make language technology work well everywhere in the world. |
Keywords
* Artificial intelligence * Precision * Quantization
