Summary of How to Train Text Summarization Model with Weak Supervisions, by Yanbo Wang et al.
How to Train Text Summarization Model with Weak Supervisions
by Yanbo Wang, Wenyu Chen, Shimin Shan
First submitted to arxiv on: 27 Aug 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 proposes a novel method for learning complex tasks using noisy or inexact labels. By breaking down the task into simpler subtasks and generating supervision signals for each, the authors demonstrate how to integrate these signals into a manageable learning procedure. The approach is showcased through a topic-based summarization system that leverages rich supervision signals to promote both summarization and topic relevance. Notably, the model can be trained end-to-end without any labels, achieving exceptional performance on the CNN and DailyMail datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper introduces a new way to learn complex tasks using noisy or inexact labels. It breaks down the task into simpler parts and generates supervision signals for each one. These signals are then used to train a model that can summarize topics accurately. The authors show how their approach works by creating a system that summarizes news articles based on topic relevance. Without any labeled data, the model still performs well. |
Keywords
* Artificial intelligence * Cnn * Summarization
