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

GPT-4

openai.com

Post reply on HN
Print view View original

701–710 of 1001 posts

Re: GPT-4

#701

Post by azeirah »

swerling wrote: post #608

On GPT-3.5 and now on GPT-4, I told it a line I could remember from a sonnet, and asked it to give me which sonnet it came from. It failed, and fabricated a sonnet that was a mashup of other sonnets. It seems like maybe GPT-4 is not good at knowing when it does not know something? Is this a common issue with LLMs? Also surprising (to me), it seems to give a slightly different wrong answer each time I restart the chat…

This is pure speculation on my part, but I don't think LLMs in their current form will be able to talk about what they do and don't know. No matter how large the input.

There is no introspection in their architecture. Introspection likely has to involve some form of a feedback mechanism and possibly even a "sense of self".

These coming years are going to be interesting though. For sure we are going to see experiments built on top of these recent amazing LLMs that _do_ have some form of short-term memory, feedback and introspection!

Giving these kinds of AIs a sense of identity is gonna be a strange thing to behold. Who knows what kind of properties will start to emerge

Re: GPT-4

#702

Post by hooande »

billythemaniam wrote: post #612
hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

If you had full source code that fit into the context, do you think it could reliably answer questions about the code, build unit tests, generate documentation? I ask because that is the software equivalent of what you just described.

Yes. It still can't attend meetings, collaborate on projects or set priorities. Or any of the other things programmers spend most of their time doing.

Also I'd guess that it still generally sucks at programming. Code has a lot of very similar sequences and logical patterns that can be broken, which makes it prone to hallucinating. I'd imagine that more parameters will help with this.

Re: GPT-4

#703

Post by zeofig »

albertgoeswoof wrote: post #294
Imnimo wrote: post #218

A class of problem that GPT-4 appears to still really struggle with is variants of common puzzles. For example: >Suppose I have a cabbage, a goat and a lion, and I need to get them across a river. I have a boat that can only carry myself and a single other item. I am not allowed to leave the cabbage and lion alone together, and I am not allowed to leave the lion and goat alone together. How can I safely get all three…

LLMs aren’t reasoning about the puzzle. They’re predicting the most likely text to print out, based on the input and the model/training data. If the solution is logical but unlikely (i.e. unseen in the training set and not mapped to an existing puzzle), then the probability of the puzzle answer appearing is very low.

Word. There is no other way it can be. Not to say these "AI"s aren't useful and impressive, but they have limitations.

Re: GPT-4

#704

Post by nullc »

hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

> As a professional...why not do this?

Because of confidentiality.

Re: GPT-4

#705

Post by 00F_ »

Sol- wrote: post #288

Interesting how quickly we are pushing ahead with obsoleting human cognition. It may bring many benefits, but I wonder if at some point this development should not be decided by society at large instead of a single well-funded entity that is in an arms race with its competitors. This endeavor is ultimately about replacing humanity with a more intelligent entity, after all. Might be that more humans should have a say…

every other day i am reminded about the state of AI and i feel complete despair. why do people not realize exactly what you just said, that this endeavor is ultimately about replacing humanity? what other long-term result could the concept of AI possibly have? its like the biggest mass psychosis that has ever existed. whenever i talk to people about this, they always parrot the same thing almost word for word: people will just find new, better jobs. or, you know, something about the Luddites. its mass psychosis because they refuse to acknowledge the blindingly obvious and plain fact that humans wont be hired to do anything if humans are the worst at doing literally any task. and what are the consequences of such a world? people just draw a blank. its like the MIB came up and flashed them and they just go on with their day. i think the same is true even with you. you make this comment "so it probably wont happen, oh well." as if it werent an existential threat.

Re: GPT-4

#706

Post by tildef »

justanotheratom wrote: post #697
OkGoDoIt wrote: post #533

From the livestream video, the tax part was incredibly impressive. After ingesting the entire tax code and a specific set of facts for a family and then calculating their taxes for them, it then was able to turn that all into a rhyming poem. Mind blown. Here it is in its entirety: --- In the year of twenty-eighteen, Alice and Bob, a married team, Their income combined reached new heights, As they worked hard day and…

Where can I watch the recording of the Livestream

https://www.youtube.com/live/outcGtbnMuQ

Re: GPT-4

#708

Post by Madmallard »

hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

GPT is censored with respect to medical diagnosis

Re: GPT-4

#709

Post by nl »

pstorm wrote: post #609
hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

Reading the press release, my jaw dropped when I saw 32k. The workaround using a vector database and embeddings will soon be obsolete.

> The workaround using a vector database and embeddings will soon be obsolete.

This is 100% not the case. Eg I use a vector database of embedding to store an embedding of every video frame which I later use for matching.

There are many NLP-only related tasks this helps for but equally as many that still require lookup and retrieval.

Re: GPT-4

#710

Post by zamnos »

braza wrote: post #455

I am glad for the OpenAI team for such advancement and how fast they integrated with several other partners (Microsoft, Duolingo); but at the same time I think the “regular” academia (ie universities and research institutes) lost the train for this kind of research (some can call academic engineering). I know that the academia is doing a great job in AI with base research (eg Stable Diffusion) but seeing those new pl…

Except that Stable Diffusion only came about because of Stability.ai and Emad's sponsorship, so I don't know that I'd use that as an example of a success by academia. It's true that the people who made it are academics, but that's to say they weren't hedge fund managers with a couple hundred thousand dollars to burn on a GPU cluster. The government and by extension its people needs to want to throw a lot more money at open ended research if we want science to be able to progress at the hands of academics and not corporations.
Post reply on HN