Summary of Cpapers: a Dataset Of Situated and Multimodal Interactive Conversations in Scientific Papers, by Anirudh Sundar et al.
cPAPERS: A Dataset of Situated and Multimodal Interactive Conversations in Scientific Papers
by Anirudh Sundar, Jin Xu, William Gay, Christopher Richardson, Larry Heck
First submitted to arxiv on: 12 Jun 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 A novel area of research in situated and multimodal interactive conversations (SIMMC) explores interactions within scientific papers. To facilitate depth of inquiry, SIMMC methods must be tailored for each paper component, including text, equations, figures, and tables. This work introduces Conversational Papers (cPAPERS), a dataset of conversational question-answer pairs derived from reviews of academic papers, grounded in these components and their associated references from arXiv documents. The study presents a data collection strategy to gather these question-answer pairs from OpenReview and associate them with contextual information from LaTeX source files. Furthermore, the authors propose baseline approaches utilizing Large Language Models (LLMs) in both zero-shot and fine-tuned configurations to address the cPAPERS dataset. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Scientific papers are usually full of text, numbers, pictures, and tables. To help people interact with these papers more effectively, a new area of research is focusing on conversations within scientific papers. This work creates a special kind of data called Conversational Papers (cPAPERS), which contains questions and answers from reviews of academic papers. The authors also explain how they collected this data and associated it with information about the papers themselves. Additionally, they test some simple language models to see how well they can answer these questions. |
Keywords
» Artificial intelligence » Zero shot
