Game World!
Join A World Of Gamers
Die Besten Rezepte Aus Der Suppen Und Eintöpfe In Deutschland
- Suppen Und Eintöpfe In Deutschland
- Liste Für 40 Kuchen Rezept In Deutschland
- How To Get Rich In 2018
- Facebook Mentorship For Business
- Einfache Rezepte Für Milcheis, Parfaits und Eis Am Stiel
- Best Practice On How To Get High Paid English Speaking Jobs In Germany
- 4 Tips To Increase Your Small Business Sales Online For Free
- 10 Interesting Fantastic Facts About Africa
Followers
Popular Posts
-
I will advice that before you do any thing on the internet trying to make money you should take some time and read this post. ...
-
Reginald Leigh Dugmore (20 November 1891 – 16 June 1967), better known as Reginald Denny , achieved success both as an English stage , f...
-
As you all know Germany is not an English speaking country. We speak here German. You happen to be an English speaking person in Germa...
-
When is gta 1000000 coming out ? Does gta 1000000 exist ? Grand Theft Auto: San Andreas is an action-adventure video game de...
-
50 ACTION STARS Then and Now | Real Name and Age 2019 80s MUSIC STARS Then and Now 70s MUSIC STARS Then and Now Micha...
-
How I lost 32 pounds of FAT and 10 inches off my waist For a lot of people, making conscious choices in the name of health requires a lot o...
-
A blog is a discussion or informational website published on the World Wide Webconsisting of discrete, often informal diary-style text e...
-
A freelancer or freelance worker , is a term commonly used for a person who is self-employed and is not necessarily committed to a part...
-
The provider says it has reset the passcodes of the current account holders whose data was compromised as it investigates the leak, the late...
-
Search engine marketing ( SEM ) is a form of Internet marketing that involves the promotion of websites by increasing their visibility in ...
Thursday 15 December 2022
Show HN: Natural language Twitter search using Codex https://ift.tt/OvPKaYL
Show HN: Natural language Twitter search using Codex We built a structured search engine for Twitter called Bird SQL, available at https://ift.tt/5LEFDO9 . Our search interface uses OpenAI Codex to translate natural language to SQL. Our backend then verifies the SQL, executes it, and displays the results on the web app. This makes large structured datasets like a scrape of Twitter easy for anyone to explore. As background, while working on text-to-SQL as a general problem, we came to believe one of its most powerful applications is as a search tool because: - SQL is hard to write by hand and prone to errors - It allows you to iterate quickly if you’re exploring a new dataset - A lot of contextual information that you’d normally have to internalize (e.g. your data’s schema) can be automatically generated and offloaded to the language model Using large language models (LLMs) like Codex to write the SQL for you means you don’t have to worry about the nitty gritty language details, but still benefit from the power of a language like SQL. Also, after seeing the results of the query, you can inspect (and if necessary, change) the SQL. The lack of this sort of explainability of the query result is one of the more notorious challenges of returning the output of an LLM directly to the user. Additionally, using LLMs in this way makes these kinds of queries over structured data accessible to people who know little or no SQL. While Bird SQL shares significant infrastructure with our more general LLM-powered search engine over unstructured data (Ask Perplexity - https://perplexity.ai[1] ), the two approaches and their respective challenges are quite different. For example, the type of models are different (GPT3.5 vs Codex), obviously the model prompts have different structure, and how to verify model output when its text vs when its code is different. We are currently exploring ways to combine the two approaches, such as using the results of retrieving information from a structured source (as in Bird SQL) as one of the inputs for the LLM to interpret or summarize (as in Ask Perplexity). We would love to hear your questions, suggestions, and feedback! [1] https://ift.tt/How0uvM https://ift.tt/PUBksa6 December 15, 2022 at 11:12PM