Summary of Increasing Faithfulness in Human-human Dialog Summarization with Spoken Language Understanding Tasks, by Eunice Akani et al.
Increasing faithfulness in human-human dialog summarization with Spoken Language Understanding tasks
by Eunice Akani, Benoit Favre, Frederic Bechet, Romain Gemignani
First submitted to arxiv on: 16 Sep 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 study in dialogue summarization tackles the challenge of accurately and faithfully summarizing conversations between multiple speakers. The researchers suggest using semantic information for Spoken Language Understanding (SLU) in human-machine dialogue systems to obtain a more semantically faithful summary regarding the task. This study introduces three key contributions: incorporating task-related information, a new evaluation criterion based on task semantics, and a new dataset version with increased annotated data standardized for research on task-oriented dialogue summarization. The evaluation uses the DECODA corpus, a collection of French spoken dialogues from a call center. Results show that integrating models with task-related information improves summary accuracy, even with varying word error rates. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study wants to make it easier to summarize conversations between multiple people. Right now, summarizing dialogue accurately is hard because you need to understand how the speakers are interacting and what’s important. Some machines can already summarize dialogue, but they might not always get it right. The researchers want to use special information about spoken language understanding to help machines summarize more accurately. They’re also introducing new ways to evaluate how well these summaries are doing and a bigger dataset with more examples to practice on. They tested their ideas using conversations from a call center and found that it works better when you take into account what the conversation is about. |
Keywords
» Artificial intelligence » Language understanding » Semantics » Summarization
