Summary of Itbls: a Dataset Of Interactive Conversations Over Tabular Information, by Anirudh Sundar et al.
iTBLS: A Dataset of Interactive Conversations Over Tabular Information
by Anirudh Sundar, Christopher Richardson, William Gay, Larry Heck
First submitted to arxiv on: 19 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 This paper introduces Interactive Tables (iTBLS), a dataset of interactive conversations situated in tables from scientific articles. The dataset is designed to facilitate human-AI collaborative problem-solving through AI-powered multi-task tabular capabilities. In contrast to prior work that models interactions as factoid QA or procedure synthesis, iTBLS broadens the scope of interactions to include mathematical reasoning, natural language manipulation, and expansion of existing tables from natural language conversation by delineating interactions into one of three tasks: interpretation, modification, or generation. The paper presents a suite of baseline approaches to iTBLS, utilizing zero-shot prompting and parameter-efficient fine-tuning for different computing situations. Additionally, it introduces a novel multi-step approach that outperforms standard parameter-efficient fine-tuning by up to 15% on interpretation, 18% on modification, and 38% on generation. The paper’s contributions include the introduction of iTBLS, a suite of baseline approaches, and a novel multi-step approach for interactive table-based AI applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new dataset called Interactive Tables (iTBLS) that helps humans and artificial intelligence work together to solve problems. The data is from scientific articles and includes conversations about tables in those articles. This is different from previous datasets that only looked at simple questions or procedures. The researchers also came up with some ways for AI to learn how to work with this new dataset. They tested these methods and found one that works really well, even better than others they tried! This paper is important because it helps us understand how humans and AI can work together more effectively. This could lead to new breakthroughs in science and other areas. |
Keywords
* Artificial intelligence * Fine tuning * Multi task * Parameter efficient * Prompting * Zero shot
