testzer0/AmbiQT
Code and Assets for "Benchmarking and Improving Text-to-SQL Generation Under Ambiguity" (EMNLP 2023)
When you ask a database a question in plain English, it sometimes has multiple valid interpretations, but most systems only give you one. This project provides a new way to get all the plausible database queries from an ambiguous English question, so you don't miss important insights. It takes an English question as input and provides multiple SQL queries as output, making it useful for data analysts or business users exploring data.
No commits in the last 6 months.
Use this if you need to ensure your natural language queries to a database capture all possible interpretations, especially when dealing with ambiguous phrasing or similar-sounding data labels.
Not ideal if your queries are always clear-cut with only one possible SQL translation, or if you primarily work directly with SQL and do not use natural language interfaces.
Stars
9
Forks
1
Language
Python
License
MIT
Category
Last pushed
Oct 15, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/testzer0/AmbiQT"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ymcui/cmrc2018
A Span-Extraction Dataset for Chinese Machine Reading Comprehension (CMRC 2018)
thunlp/MultiRD
Code and data of the AAAI-20 paper "Multi-channel Reverse Dictionary Model"
princeton-nlp/DensePhrases
[ACL 2021] Learning Dense Representations of Phrases at Scale; EMNLP'2021: Phrase Retrieval...
IndexFziQ/KMRC-Papers
A list of recent papers regarding knowledge-based machine reading comprehension.
danqi/rc-cnn-dailymail
CNN/Daily Mail Reading Comprehension Task