Summary of High-dimension Human Value Representation in Large Language Models, by Samuel Cahyawijaya et al.
High-Dimension Human Value Representation in Large Language Models
by Samuel Cahyawijaya, Delong Chen, Yejin Bang, Leila Khalatbari, Bryan Wilie, Ziwei Ji, Etsuko Ishii, Pascale Fung
First submitted to arxiv on: 11 Apr 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 proposed UniVaR framework aims to align Large Language Models (LLMs) with human values and preferences without requiring a large-scale human annotation effort. To achieve this, UniVaR represents high-dimensional human value distributions in LLMs, orthogonal to model architecture and training data. The authors trained UniVaR using the value-relevant output of eight multilingual LLMs and tested it on four LLMs (LlaMA2, ChatGPT, JAIS, and Yi). The results show that UniVaR is a powerful tool for comparing human values embedded in different LLMs from various language sources. This framework sheds light on the complex interplay between human values and language modeling. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us understand how large language models prioritize different values in different languages and cultures. It’s like finding out what kind of values are important to each model, and how those values change depending on the language it was trained on. The authors developed a special tool called UniVaR that can compare these values across different models and languages. |
