Summary of Hallucination Diversity-aware Active Learning For Text Summarization, by Yu Xia et al.
Hallucination Diversity-Aware Active Learning for Text Summarization
by Yu Xia, Xu Liu, Tong Yu, Sungchul Kim, Ryan A. Rossi, Anup Rao, Tung Mai, Shuai Li
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); 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 The proposed active learning framework, HAllucination Diversity-Aware Sampling (HADAS), aims to alleviate Large Language Models’ (LLMs) propensity for generating factually incorrect or unsupported outputs. By measuring fine-grained hallucinations from errors in semantic frame, discourse and content verifiability, the approach selects diverse hallucinations for annotations in active learning for LLM finetuning. The framework is demonstrated to be effective and efficient in mitigating LLM hallucinations on three datasets and different backbone models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper proposes a new way to help Large Language Models stop making mistakes. When these models generate text, they sometimes produce facts that are not true or do not make sense. Existing methods for fixing this problem require humans to carefully check and correct the mistakes. However, most of these methods only work for specific types of mistakes. This paper proposes a new approach that can fix different kinds of mistakes in one go. |
Keywords
* Artificial intelligence * Active learning * Discourse * Hallucination
