Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
blazespin
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
91.
▲
Graph of thoughts, BFS/DFS of LLM reasoning chains
(github.com)
1 points
by
blazespin
3y ago
|
0 comments
92.
▲
by
blazespin
3y ago
The dictionary definition of reasoning says nothing about how the thinking is done, only that it's sensible and logical, which is exactly what GPT4 is. Limited, yes, but it reasons.
93.
▲
by
blazespin
3y ago
The resolution is actually fairly simple. It's an incredibly brilliant stochastic parrot with some limited reasoning capabilities. Some folks will try to say it cannot reason, but they are wrong, there is extensive proof of that. The
94.
▲
by
blazespin
3y ago
The sequence of these two threads is just too perfect. Almost likely someone is trying to make a point.
95.
▲
by
blazespin
3y ago
And another - https://www.youtube.com/watch?v=MIfhunAwZew These folks have been working on this for quite awhile.
96.
▲
by
blazespin
3y ago
Here's a similar project that was done awhile ago https://github.com/qrdlgit/simbiotico
97.
▲
by
blazespin
3y ago
No, the opposite actually. GPT4 enabled this.
98.
▲
by
blazespin
3y ago
We actually submitted a patent on this exact idea (using hierarchical list centric design for ideation with LLMs) Maybe not as non obvious as we thought! :)
99.
▲
Improving Small Language Models in Domain-Specific QA
(arxiv.org)
2 points
by
blazespin
3y ago
|
0 comments
100.
▲
Should We Tax Robots?
(news.mit.edu)
40 points
by
blazespin
3y ago
|
73 comments
101.
▲
by
blazespin
3y ago
GPT4 struggles with things for sure, but this isn't one of them. You're just projecting what you think should be the right answer. If it worries you that much, try adding "Provide alternative responses to this question. E
102.
▲
by
blazespin
3y ago
Interestingly, I find ChatGPT very reticent about giving out PII of anyone not notable. This might not be a huge issue
103.
▲
by
blazespin
3y ago
You'll want to fact check this, but I find GPT4 is OK at this kinda stuff. If you've been asked to remove data in accordance to a gdpr request, is it best effort or do you have to strictly comply with the request? ChatGPT I am not
104.
▲
by
blazespin
3y ago
Yeah, it's like a month I believe.
105.
▲
by
blazespin
3y ago
I think you'd have to be a GDPR lawyer to understand the implications of that. It can get a little complicated.
106.
▲
by
blazespin
3y ago
Most likely a post filter. Unfortunately for OpenAI and anyone creating something similar, it's probably hackable. Not sure how best efforts work with GDPR.
107.
▲
by
blazespin
3y ago
Yeah, unless you're a shop running outlook and teams, not sure you really have a say here. Maybe this is what customers want. Also, not entirely clear that verge isn't just knee jerk reporting. Any sys admins that have actual fi
108.
▲
by
blazespin
3y ago
Verge seems to be only one covering this. Any sysadmins here with 1st hand experience with the change?
109.
▲
by
blazespin
3y ago
It's absolutely ludicrous nonsense. It's like saying someone invented the idea of 'wondering what would happen if the sun blew up'. VR is a trivial idea that pretty much everyone has thought about at some point. It
110.
▲
by
blazespin
3y ago
pretty sure first person that had a dream figured out VR.
111.
▲
by
blazespin
3y ago
"who first conceived the idea of virtual reality"
112.
▲
by
blazespin
3y ago
Just another poor person in a stage of grief. LLMs are effective at both memorization and reasoning. They are inevitable in the same way trains were inevitable. Let's just figure out how best to use these and move forward as we alwa
113.
▲
by
blazespin
4y ago
I think there are FTC laws around this maybe.
114.
▲
by
blazespin
4y ago
The fact is nobody can risk not owning a piece of the foundational models. There is waaaay too much upside risk that they will tots dominate the market. I mean, maybe they won't like you say, but what if they do? Then you're prob
115.
▲
by
blazespin
4y ago
Sure, if you understand the underlying point of that quote - if you have to say it, you're probably thinking about the person.
116.
▲
by
blazespin
4y ago
Well, I stopped using Google personally, except for online docs.
117.
▲
by
blazespin
4y ago
Not to mention the ludicrously gaping security issue that this is. My guess is they want to push people to the plugins tho.
118.
▲
by
blazespin
4y ago
It's not quite there yet, but it certainly could be.
119.
▲
by
blazespin
4y ago
I dunno, the stock price hasn't really tanked, so it didn't really backfire that much and it allowed them to get a bit of feedback as they try to figure out how to maintain revenue. What if they put out this insanely great model a
120.
▲
by
blazespin
4y ago
Google is in a weird spot. I suspect they are capable of doing so much more, but there is a serious risk to cannibalizing their 99% revenue model (search), which they probably don't yet know if they can monetize in the same way. Unfor
More ›