101–107 of 107 posts
Earlier quoted context omitted.
You basically just described a GAN. Neat!
GANs work by feeding back the mistakes and forcing the generator model to improve its cheating. In this case, filtering out titles that are ambiguous would act as an independent filter.
Earlier quoted context omitted.
"The structure of the HIV capsid is analysed by cryo-electron microscopy and cryo-electron microscopy at cryo-electron-microscopy resolution." It really does like to repeat itself.
This one made me laugh really really hard: "This study presents the phylogenetic characterization of the beak and beak of beak whales; it is suggested that the beak and beak-toed beaks share common cranial bones, providing support for the idea that beaks are a new species of eutriconodont mammal."
[I know nothing - I'm pretty ignorant about practical ML]
Earlier quoted context omitted.
I find the whole thing ominous because there is no "there" there: there is no understanding in the GPT-2 system, but it's able to generate increasingly plausible text. This greatly increases the amount of plausible nonsense that can be used to drown out actual research. You could certainly replace a lot of pop-sci and start several political movements with GPT-2... all of which has no actual nutritional content.
No, the text is generally very implausible if you know anything about the science. For example, describing dna twists with protein folding descriptors, mixing up quantum computing with astronomy, inorganic chemistry with biochemistry, virology with bacteriology... I was really impressed with gpt-2 but seeing this really gave me a feel for how much of a lack of understanding it has.
[KR²] https://en.wikipedia.org/wiki/Knowledge_representation_and_r...
Cool demo. With these GPT models, I don't get the appeal of creating fake text that at best can pass as real to someone who doesn't understand the topic and context. What's the use case? Generating more believable spam for social media? Anything else? Because there's no real knowledge representation or information extraction going on here.
I use GPT-2/3 in creative writing to generate rough text that I then go back and edit/improve because it gives a good starting point (and often has a lot of high-quality factors). I don't know if this translates to technical writing, but it's possible someone might complete a prompt on some specific topic(s) and then use that as a point to start from, especially if they're knowledgeable enough on the topic to correct…
How does one train GPT-2 with their own content and produce nice results at arbitrary lengths? I found a few libraries but I could not use them well, I get lost very quickly. I just want to train our internal Confluence and have fun with it.
If we feed AI all the knowledge about the physics of the world, then will it be ever capable of giving answers without actually performing scientific research inferring it just from the laws that define the world?
On the other hand, an interesting possibility with well-designed text-mining and AI models would be for them to generate valid hypotheses that hadn't been contemplated earlier, based on the massive corpus of scientific publications. The model may be able to find possible correlations or interesting ideas by combining sources from different fields that would normally be ignored by the over-specialised research community. However, in that case the model wouldn't be valuable for providing answers—rather, it's value would be in providing questions.
Earlier quoted context omitted.
I use GPT-2/3 in creative writing to generate rough text that I then go back and edit/improve because it gives a good starting point (and often has a lot of high-quality factors). I don't know if this translates to technical writing, but it's possible someone might complete a prompt on some specific topic(s) and then use that as a point to start from, especially if they're knowledgeable enough on the topic to correct…
What platform do you use ? Is it all run from home ?
I also built a more polished version to add to the Notebook.ai document editor (so writers can get some continuation prompts whenever they get a bit of writer's block), but the pricing made it unfeasible to actually release. Notebook.ai is also open source though, and you can see the GPT-3 functionality in the unmerged PR here: https://github.com/indentlabs/notebook/pull/739