Well, sometimes I have this "Google Duplex: A.I. Assistant Calls Local Businesses To Make Appointments" feeling.
Gemini "duck" demo was not done in realtime or with voice
551–560 of 683 posts
Re: Gemini "duck" demo was not done in realtime or with voice
#552Earlier quoted context omitted.
It's been 3.5 years since GPT-3 was released, and just over a year since ChatGPT was released to the public. If it was possible to solve LLM hallucinations with simple Chain-of-Thought style agents, someone would have done that and released a product by now. The fact that nobody has released such a product, is pretty strong evidence that you can't fix hallucinations via Chain-of-Thought or Retrieval-Augmented Generat…
I agree: but I just wanted to say that there are specific subdomains where you can mitigate some of these issues. For example, generating json. You can explicitly follow a defined grammar to get what will always be a valid json output. Similarly, structured output such as code can be passed to other tools such as compilers, type checkers and test suites to ensure that at a minimum the output you selected passes some…
I agree that the "forcing valid json output" is super cool.
But it's unrelated to the problem of LLM hallucinations. A hallucination that's been validated as correct json is still a hallucination.
And if your problem space is simple enough that you can validate the output of an LLM well enough to prove it's free of hallucinations, then your problem space doesn't need an LLM to solve it.
Re: Gemini "duck" demo was not done in realtime or with voice
#553While this might just be a bit of bad PR now, it will eventually be a nothing burger. Remember the original debut of Apple's Siri for which Apple also put out a promotional demo with greatly exaggerated functionality? People even sued Apple and they lost.
As much as I hate it, this is absolutely fine by our society's standards. https://www.theregister.com/AMP/2014/02/14/apple_prevails_in...
Re: Gemini "duck" demo was not done in realtime or with voice
#554Is the link to the article broken and anyone has it archived somewhere? I wish people stop posting Twitter messages to HN and provide a link directly to the original article. What's next, post on HN on an Instagram post?
I dont get it too, my browser was loading for good 15 seconds and made 141 requests fetching almost 9 MB of resources to show me exactly same content as provided in OpenGraph tags and a freaking redirect to a Bloomberg link. Feels like a slap in the face to open such phishing link at any time, just a useless redirect with nine million bytes of overhead.
Re: Gemini "duck" demo was not done in realtime or with voice
#555Earlier quoted context omitted.
Is this cheating? It sounds like cheating and reflects quite poorly on you.
> It was our first comp sci class ever, we were given raspberry pi's. We had no coding experience or guidance, and were asked to create "something". Garbage in, garbage out.
More like cheaters in, cheaters out.
Re: Gemini "duck" demo was not done in realtime or with voice
#556A big red flag for me was that Sundar was prompting the model to report lots of facts that can be either true or false. We all saw the benchmark figures that they published and the results mostly showed marginal improvements. In other words, the issue of hallucination has not been solved. But the demo seemed to imply that it had. My conclusion was that they had mostly cherry picked instances in which the model happen…
i think this was demonstrated in that mark rober promo video[1] where he asked why the paper airplane stalled by blatantly leading the witness. "do you believe that a pocket of hot air would lead to lower air pressure causing my plane to stall?" he could barely even phrase the question correctly because it was so awkward. just embarrassing. [1] https://www.youtube.com/watch?v=mHZSrtl4zX0&t=277s
This has got to be satire! That is too funny.
Re: Gemini "duck" demo was not done in realtime or with voice
#557I wanted to see what's really going on, but the bloomberg article on the twit link seems taken down right now.
Re: Gemini "duck" demo was not done in realtime or with voice
#558This is so crazy. Google invented transformers which is the bases for all these models. How do they keep fumbling like this over and over. Google Docs created in 2006! Microsoft is eating their lunch. Google creates the ability to change VM's in place and makes a fully automated datacenter. Amazon and Microsoft are killing them in the cloud. Google has been working on self driving longer than anyone. Tesla is catchin…
I was at MS in 2008 September and internally they had a very beautiful and well functioning Office web already (named differently, forgot the name but it wasn't sharepoint if I recall correctly, I think it had to do something with expense reports?) that would put Google Docs to shame today. They just didn't want to cannibalize their own product.
>I was at MS in 2008 September and internally they had a very beautiful and well functioning Office web already
So why did they never release that and went with Office 365 instead?
Re: Gemini "duck" demo was not done in realtime or with voice
#559Earlier quoted context omitted.
> It was our first comp sci class ever, we were given raspberry pi's. We had no coding experience or guidance, and were asked to create "something". Garbage in, garbage out.
Wow this is such an awful excuse. Here’s a whole list of projects intended for kids. https://all3dp.com/2/best-raspberry-pi-projects-for-kids/ It includes building out a whole weather station which includes a humidity sensor as one of the many things it can do.
> Wow this is such an awful excuse.
yes for whomever organized such a curse and didn't give such guidance.
And besides curse asked for project to do something. It did. It printed lines. We can call the email gimmick, the marketeering strategy, making a turd look good.
Don't blame students for failure of whomever designed the curse.
Re: Gemini "duck" demo was not done in realtime or with voice
#560Earlier quoted context omitted.
I imagine the commenter was calling out what they perceived to be an inauthentic yet carefully planned facade of diversity. This marketing trend rubs me the wrong way as well, because it reminds me of how I was raised and educated as a 90s kid to believe that racism was a thing of the past. That turned out to be a damaging lie. I don't mean to imply that companies should avoid displays of diversity, I just mean that…
You'd prefer the alternative with just a few white guys in the picture and no consideration given at all to appearing diverse?
Just take a group of people that actually know and work together and you're authentic. Forced diversity is idiotic: either you do it or you don't, but you show what you're doing to be authentic.
Imagine how cringe it would be if only white guys were allowed to work at Google and they displayed in all their marketing a fully diverse group of non-white girls. That would be... inauthentic.
Just the fact girls are less than guys in IT is something we should demonstrate, understand, change if needed. Not hide behind a facade of 50/50 display everywhere as if the problem was already solved or that it was even a problem in the first place.