[HN Gopher] Improving the factual accuracy of language models th...
___________________________________________________________________
Improving the factual accuracy of language models through web
browsing
Author : grappler
Score : 38 points
Date : 2021-12-16 17:27 UTC (5 hours ago)
(HTM) web link (openai.com)
(TXT) w3m dump (openai.com)
| vimax wrote:
| The next version of copilot will literally be a bot that searches
| Google and reads through stack overflow posts. There are a lot of
| jokes you can make, but that is honestly what I do half the day
| and I've met some people who could barely do that.
|
| First they came for the junior dev, but I was not concerned for I
| was not a junior dev.
| TaylorAlexander wrote:
| I like the note that there is a new risk since their model can
| access the web at training time. A model in this situation could
| potentially do something malicious online!
| gwern wrote:
| "Don't worry, we'll just make sure it can only do GET commands,
| which are, per the HTTP spec, side-effect free and idempotent."
|
| [later]
|
| "Oh no."
|
| ---
|
| Ha-ha-only-serious joking aside, this is impressive. Early on
| in the API release, someone showed you could _sort of_ make
| GPT-3 operate on text representations to put inside a web loop,
| but it made a lot of errors. This goes well beyond 'sort of'.
| It may not seem like 'predict the next word' gets you far, but
| turns out, it's a good starting point. ("Tool AIs want to be
| agent AIs.")
| skybrian wrote:
| I was wondering whether it obeys robots.txt? I guess since it's
| using Bing, it obeys as well as Bing does?
___________________________________________________________________
(page generated 2021-12-16 23:01 UTC)