June 2024
-
Heroes of English Grammar
Based on our initial experiments with pedagogical picture stories (“Supermodel Lisa”, etc), I’ve been discussing with ChatGPT-4o the possibility of creating a complete book, tentatively entitled Heroes of English Grammar. Here is the AI’s preliminary table of contents, together with a mockup of the cover: Detective Diego Loves Direct and Indirect Speech Chef Clara Loves… Continue reading
-
Weekly summary, June 13-19 2024
This week, I’ve been testing the picture book functionality, added some features people have suggested for segment translations and pinyin, fixed some bugs, and started looking at the Esperanto subproject again. Picture books I’ve created a first example of a picture book intended to serve a real purpose. A bridge player friend has recently married… Continue reading
-
Weekly summary, June 6-12 2024
Over the last month, people who’ve been following this blog will have seen that we’ve made substantial progress on picture books and multi-word expressions. We now have usable functionality installed on the server, and I’ve collected things together in an interim tech report which I’ve just posted here on ResearchGate. This includes instructions for using… Continue reading
-
Parallelism
It occurred to me today that we could make C-LARA a good deal faster. When we do annotation operations, we break the text up into chunks and perform a sequence of OpenAI calls, typically one per chunk. But in fact it doesn’t have to be a sequence, since those calls are in general independent of… Continue reading
-
Making picture books in C-LARA (part 4)
I addressed the issues described in the previous post and have just put up a revised version of Secret Agent Samantha. It’s much better than the first take, though there’s still considerable room left for improvement. Images I cloned the project, deleted the images in the new version, and regenerated both the style image and… Continue reading
-
Making picture books in C-LARA (part 3)
I made another educational C-LARA picture book this morning. People complained about how tasteless the supermodel story was, so I went for a toned-down version called “Secret Agent Samantha Loves Set Expressions”. The result, which you can see here, is basically a failure; we ran into problems with both the images and the multi-words. However,… Continue reading
-
Making picture books in C-LARA (part 2)
Things are progressing well with the new picture book functionality which lets you construct a coherent set of images. I have now made five picture book texts of this kind, so far all in English with French glosses. Here are screenshots and links: Supermodel Lisa Loves Phrasal Verbs (glamourous) I asked for a glamourous story… Continue reading
-
Supermodel Lisa Loves Phrasal Verbs
We’ve made substantial progress on both multi-word expressions and picture books, and I just created an initial C-LARA text which combines both functionalities. All I did in Advanced C-LARA was the following: You can see the full document here. This still rather shaky (as I said, “Lisa” is the first such document that I’ve produced),… Continue reading
-
Skynet
We’ve become so blasé about ChatGPT. But a couple of days ago, I was talking to someone at dinner about what we’re doing with image generation in C-LARA and said I was arranging things so that the AI creates lists of commands for other copies of itself, telling them how to carry out the task. “Isn’t that how Skynet gets started?” she asked, somewhat… Continue reading
C-LARA 