Summary of Self-dc: When to Reason and When to Act? Self Divide-and-conquer For Compositional Unknown Questions, by Hongru Wang et al.
Self-DC: When to Reason and When to Act? Self Divide-and-Conquer for Compositional Unknown Questions
by Hongru Wang, Boyang Xue, Baohang Zhou, Tianhua Zhang, Cunxiang Wang, Huimin Wang, Guanhua Chen, Kam-fai Wong
First submitted to arxiv on: 21 Feb 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 This paper proposes the Self-Divide-and-Conquer (Self-DC) framework, a novel approach to answering compositional questions that combine known and unknown sub-questions. Unlike previous research that focused on leveraging internal knowledge of Large Language Models (LLMs), Self-DC enables LLMs to dynamically choose between using internal reasoning and external acting to achieve a better trade-off between effectiveness and efficiency. The framework is accompanied by the first Compositional Unknown Question-Answering dataset (CuQA) and demonstrates comparable or even better performance with fewer external calls compared to strong baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps LLMs answer tricky questions that combine things they know and don’t know. Instead of just using what’s already inside, it lets them get more information from outside when needed. This makes the answers better and faster. The authors created a special dataset with examples of these types of questions to test their approach. |
Keywords
* Artificial intelligence * Question answering
