{"id":2117,"date":"2021-01-06T12:49:39","date_gmt":"2021-01-06T12:49:39","guid":{"rendered":"https:\/\/thenextweb.com\/?p=1333166"},"modified":"2021-01-06T12:49:39","modified_gmt":"2021-01-06T12:49:39","slug":"say-hello-to-openais-dall-e-a-gpt-3-powered-bot-that-creates-weird-images-from-text","status":"publish","type":"post","link":"https:\/\/www.londonchiropracter.com\/?p=2117","title":{"rendered":"Say hello to OpenAI\u2019s DALL-E, a GPT-3-powered bot that creates weird images from text"},"content":{"rendered":"\n

OpenAI<\/span> has unveiled a neural network known as DALL-E that converts text into striking images \u2014 like these pictures of a baby panda in a tutu wielding a lightsaber:<\/p>\n

Credit: OpenAI<\/figcaption>
<\/i><\/a>Cute \u2014 but deadly.<\/figcaption><\/figure>\n

OpenAI said in a blogpost<\/a> that the system is a 12 billion parameter version of the landmark GPT-3 language model:<\/p>\n

\n

We\u2019ve found that it has a diverse set of capabilities, including creating anthropomorphized versions of animals and objects, combining unrelated concepts in plausible ways, rendering text, and applying transformations to existing images.<\/p>\n<\/blockquote>\n

DALL-E \u2014 a portmanteau of the names of surrealist artist Salvador Dal\u00ed and Pixar\u2019s robot WALL-E \u2014 was trained on a dataset of text-image pairs extracted from the internet.<\/p>\n

This allows it to create entirely new pictures by exploring the structure of a prompt \u2014 including fantastical objects combining unrelated ideas that it was never fed in training.<\/p>\n

[Read: Meet the 4 scale-ups using data to save the planet<\/a>]<\/em><\/p>\n

It can produce some seriously impressive images of landmarks, locations, hybrid animals, and designs from different decades. But OpenAI admits that not all the results are successful.<\/p>\n

Credit: OpenAI<\/figcaption>
<\/i><\/a>Some locations \u201cevoke a sense of d\u00e9j\u00e0 vu\u2014eerie simulacra of streets, sidewalks, and cafes that remind us of very specific locations that do not exist.\u201d<\/figcaption><\/figure>\n

The company said the system sometimes fails to draw some of the specified items and confuses the associations between objects and their specified attributes:<\/span><\/p>\n

\n

Generally, the longer the string that DALL-E is prompted to write, the lower the success rate. We find that the success rate improves when parts of the caption are repeated. Additionally, the success rate sometimes improves as the sampling temperature for the image is decreased, although the samples become simpler and less realistic.<\/p>\n<\/blockquote>\n

DALL-E also often reflects superficial stereotypes when answering queries about geographical facts, such as flags, cuisines, and local wildlife. This shortcoming is particularly significant in light of the growing concerns about the biases of large-language models<\/a>.<\/p>\n

OpenAI said it plans to analyze the societal impacts of models such as DALL-E. But the company believes the system shows \u201cthat manipulating visual concepts through language is now within reach.\u201d<\/p>\n

You can try out a controlled demo of DALL-E for yourself at the OpenAI website<\/a>.<\/p>\n

Published January 6, 2021 \u2014 12:49 UTC <\/p>\n

Source<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

OpenAI has unveiled a neural network known as DALL-E that converts text into striking images \u2014 like these pictures of a baby panda in a tutu wielding a lightsaber: Credit: OpenAICute \u2014…<\/p>\n","protected":false},"author":1,"featured_media":2118,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"_links":{"self":[{"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=\/wp\/v2\/posts\/2117"}],"collection":[{"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2117"}],"version-history":[{"count":0,"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=\/wp\/v2\/posts\/2117\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=\/wp\/v2\/media\/2118"}],"wp:attachment":[{"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2117"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2117"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.londonchiropracter.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2117"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}