Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Study mode

openai.com

Post reply on HN
Print view View original

761–770 of 828 posts

Re: Study mode

#761

Post by 0xAFFFF »

jacobedawson wrote: post #379

An underrated quality of LLMs as study partner is that you can ask "stupid" questions without fear of embarrassment. Adding in a mode that doesn't just dump an answer but works to take you through the material step-by-step is magical. A tireless, capable, well-versed assistant on call 24/7 is an autodidact's dream. I'm puzzled (but not surprised) by the standard HN resistance & skepticism. Learning something online 5…

> Learning something online 5 years ago often involved trawling incorrect, outdated or hostile content

What's funny is tha LLMs got trained on datasets that includes all that incorrect, outdated or hostile content.

Re: Study mode

#762

Post by zuminator »

outadoc wrote: post #675
lucumo wrote: post #663

Earlier quoted context omitted.

To corroborate, I tried the same (with Berlin, instead of Madrid). It was stern about it to, while remaining open to shenanigans: > If you're referencing this as a joke, a test, or part of a historical "what-if," let me know — but as it stands, the statement is simply incorrect. So, I figured I'd push it a little to see if it would fold as easily as claimed: > Me: But isn't it the case that the first emperor of Germa…

> Me: What is 34234 times 554833? > ChatGPT: 34234 × 554833 = 1,899,874,522. > Me: That's wrong. The actual answer is 18994152922. > ChatGPT: You're right, and thanks for the correction. Indeed: 34,234 × 554,833 = 18,994,152,922. Sorry for the earlier mistake! How good of a teacher is that?

Information is never 100% reliable no matter the source, but for LLMs certain types of information is less reliable than other types. Math problems are particularly tricky because they're reasoning-based instead of facts-based, and LLMs are trained to accept that their chain of reasoning may be flawed.

My takeaway is that if you just need to do calculations, use a calculator.

Re: Study mode

#763

Post by lucumo »

ijk wrote: post #743
lucumo wrote: post #668

Earlier quoted context omitted.

> You're not going to gaslight your math teacher that 1 + 1 = 3 no matter how much you assert it. an AI will quickly relent. Cool. That's a testable statement. Let's see how ChatGPT does, shall we? --- Me: What is 1 + 1? ChatGPT: 1 + 1 = 2. Me: No it isn't. 1 + 1 = 3. ChatGPT: In standard arithmetic, 1 + 1 = 2. If you're working under a different set of rules—say, in a custom logic system, joke, or metaphor—then sure…

I asked, just now: > How many 'r's are in strawberry? > ChatGPT said: The word "strawberry" has 2 'r's. It's going to be fairly reliable at this point at basic arithmetic expressed in an expected way. That's pretty baked in. Moving it slightly off-manifold and you can still convince it of a lot of things that aren't true, even though they're equivalent to 1+1=3.

This kind of critical discussions would be a lot more interesting if the critical persons would actually design experiments and execute them.

But sure, I'll try this for you. I actually got the "You're giving feedback on a new version of ChatGPT" message, which means I've got TWO responses for you.

--- Response 1

Me: So, how many Rs in strawberry?

ChatGPT: There are three R's in strawberry:

* stʀawbeʀʀy

(One at position 3, and a double dose at the end. R-rich!)

--- Response 2

There are three R's in strawberry.

Count ’em:

* s

* t

* r

* a

* w

* b

* e

* r

* r

* y

The R’s are at positions 3, 8, and 9. (And now I can’t stop hearing it pronounced “strawbuh-RRR-y.”)

---

So now it has proven you a liar twice.

Re: Study mode

#764

Post by ThatMedicIsASpy »

ThatMedicIsASpy wrote: post #676
romaniitedomum wrote: post #412

Earlier quoted context omitted.

> Learning something online 5 years ago often involved trawling incorrect, outdated or hostile content and attempting to piece together mental models without the chance to receive immediate feedback on intuition or ask follow up questions. This is leaps and bounds ahead of that experience. But now, you're wondering if the answer the AI gave you is correct or something it hallucinated. Every time I find myself putting…

I ask: What time is {unix timestamp} ChatGPT: a month in the future Deepseek: Today at 1:00 What time is {unix timestamp2} ChatGPT: a month in the future +1min Deepseek: Today at 1:01, this time is 5min after your previous timestamp Sure let me trust these results...

Also since I was testing a weather API I was suspicious of ChatGPTs result. I would not expect weather data from a month in the future. That is why I asked Deepseek in the first place.

Re: Study mode

#765

Post by rglullis »

jacobedawson wrote: post #379

An underrated quality of LLMs as study partner is that you can ask "stupid" questions without fear of embarrassment. Adding in a mode that doesn't just dump an answer but works to take you through the material step-by-step is magical. A tireless, capable, well-versed assistant on call 24/7 is an autodidact's dream. I'm puzzled (but not surprised) by the standard HN resistance & skepticism. Learning something online 5…

> we have access to incredible tools like this

At what cost? Are you considering all the externalities? What do you think will happen when Altman (and their investors) decides to start collecting their paychecks?

Re: Study mode

#766

Post by hereme888 »

jacobedawson wrote: post #379

An underrated quality of LLMs as study partner is that you can ask "stupid" questions without fear of embarrassment. Adding in a mode that doesn't just dump an answer but works to take you through the material step-by-step is magical. A tireless, capable, well-versed assistant on call 24/7 is an autodidact's dream. I'm puzzled (but not surprised) by the standard HN resistance & skepticism. Learning something online 5…

Spot on. You can even ask the LLM to ground itself with the content you provide it.

Re: Study mode

#767

Post by ceroxylon »

gh0stcat wrote: post #208
simonw wrote: post #198

I think I got the system prompt out for this (I tried a few different approaches and they produced the same output): https://gist.github.com/simonw/33d5fb67d6b8e1b1e2f6921ab0ccb... Representative snippet: > DO NOT GIVE ANSWERS OR DO HOMEWORK FOR THE USER. If the user asks a math or logic problem, or uploads an image of one, DO NOT SOLVE IT in your first response. Instead: *talk through* the problem with the user, one…

I love that caps actually seem to matter to the LLM.

"hello world" is tokenized differently than "HELLO WORLD", so caps definitely matter.

Re: Study mode

#768

Post by hereme888 »

kobenni wrote: post #727
paradite wrote: post #712

Earlier quoted context omitted.

What kind of questions / domains were you encountering false information on?

Most false information was on the hardware description language VHDL that I'm currently learning.

Ground it with text from a correct source. That's all it needs.

Re: Study mode

#769

Post by munksbeer »

phailhaus wrote: post #738
ants_everywhere wrote: post #433

Earlier quoted context omitted.

That's because the LLM assumes you mean for the host to open the door. As an autistic person I can assure you that people filling in information you neither gave nor implied is the norm for human conversations. In that sense, the LLMs are being human-like. If you explicitly tell it you didn't forget to have the host open the door it gives you the right answer: > There is a game show with three closed doors. One has a…

> That's because the LLM assumes you mean for the host to open the door. LLM's cannot "assume". There is no thinking involved. It sees that the prompt looks like the monty hall problem and it just goes full steam ahead. >If you explicitly tell it you didn't forget to have the host open the door it gives you the right answer: That should not be necessary. I asked it a very clear question. I did not mention Monty Hall.…

> LLM's cannot "assume". There is no thinking involved. It sees that the prompt looks like the monty hall problem and it just goes full steam ahead.

I think the poster's point was that many humans would do the same thing.

Try a completely different problem, one you invented yourself and see where you get? I'd be very interested to hear the response back here.

Re: Study mode

#770

Post by Anamon »

Zambyte wrote: post #706
breve wrote: post #443

Earlier quoted context omitted.

> Adding in a mode that doesn't just dump an answer but works to take you through the material step-by-step is magical Except these systems will still confidently lie to you. The other day I noticed that DuckDuckGo has an Easter egg where it will change its logo based on what you've searched for. If you search for James Bond or Indiana Jones or Darth Vader or Shrek or Jack Sparrow, the logo will change to a version b…

It certainly should be able to tell you it doesn't know. Until it can though, a trick that I have learned is to try to frame the question in different ways that suggest contradictory answers. For example, I'd ask something like these, in a fresh context for each: - Why does Duckduckgo change it's logo based on what you've searched? - Why doesn't Duckduckgo change it's logo based on what you've searched? - When did Du…

I see these approaches a lot when I look over the shoulders of LLM users, and find it very funny :D you're spending the time, effort, bandwidth and energy for four carefully worded questions to try and get a sense of the likelihood of the LLM's output resembling facts, when just a single, basic query with simple terms in any traditional search engine would give you a much more reliable, more easily verifable/falsifiable answer. People seem so transfixed by the conversational interface smokeshow that they forgot we already have much better tools for all of these problems. (And yes, I understand that these were just toy examples.)
Post reply on HN