Summary of How Good Is My Story? Towards Quantitative Metrics For Evaluating Llm-generated Xai Narratives, by Timour Ichmoukhamedov et al.
How good is my story? Towards quantitative metrics for evaluating LLM-generated XAI narratives
by Timour Ichmoukhamedov, James Hinns, David Martens
First submitted to arxiv on: 13 Dec 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 framework utilizes large language models (LLMs) to convert quantitative explanations into user-friendly narratives, enabling decision-making transparency in smaller prediction models. To evaluate the narratives without relying on human preference studies or surveys, several automated metrics are developed and applied across different datasets and prompt types to compare state-of-the-art LLMs. The framework’s utility is demonstrated by identifying new challenges related to LLM hallucinations for XAI narratives. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers has created a way to use big language models to explain the decisions made by smaller prediction models in a way that’s easy to understand. They want to find out if their method works well without asking people which explanations they like best. To do this, they came up with some automatic ways to measure how good the explanations are and tested them on different datasets and types of prompts. Their results showed that their approach can help spot new problems that arise when using these language models for explanation purposes. |
Keywords
» Artificial intelligence » Prompt
