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

An LLM is a lossy encyclopedia

simonwillison.net

Post reply on HN
Print view View original

221–230 of 365 posts

Re: An LLM is a lossy encyclopedia

#221

Post by morpheos137 »

mallowdram wrote: post #177
ACCount37 wrote: post #174

Earlier quoted context omitted.

What? This is even less coherent. You weren't talking to GPT-4o about philosophy recently, were you?

I'd know cutting-edge linguistics and signaling theory well beyond Shannon to parse this, not NLP or engineering reduction. What I've stated is extremely coherent to Systemic Functional Linguists. Beyond this point engineers actually have to know what signaling is, rather than 'information.' https://www.sciencedirect.com/science/article/abs/pii/S00033... Ultimately, engineering chose the wrong approach to automating…

If not language what training substrate do you suggest? Also not strong ideas are expressible coherently. You have an ironic pattern in your comments of getting lost in the very language morass you propose to deprecate. If we don't train models on language what do we train them on? I have some ideas of my own but I am interested if you can clearly express yours.

Re: An LLM is a lossy encyclopedia

#223

Post by s0rce »

ARandumGuy wrote: post #199
christkv wrote: post #156

Earlier quoted context omitted.

I find if I force thinking mode and then force it to search the web it’s much better.

But at that point wouldn't it be easier to just search the web yourself? Obviously that has its pitfalls too, but I don't see how adding an LLM middleman adds any benefit.

Not always, it can find stuff that is otherwise difficult for me and search engines have become much worse than 15-20 years ago.

Re: An LLM is a lossy encyclopedia

#224

Post by stego-tech »

simonw wrote: post #214
stego-tech wrote: post #173

Earlier quoted context omitted.

True, but in that case we call it “errors” or “propaganda”, depending on the context and source. Plus the steep costs of traditional encyclopedias, the need to refresh collections with new data periodically, and the role of librarians, all acted as a deterrent against lying (since they’re reference material). Wikipedia can also lie, obviously, but it at least requires sources to be cited, and I can dig deeper into to…

> Normalizing LLMs as “lossy encyclopedias” is a dangerous trend in my opinion, because it effectively handwaves the need for critical thinking skills associated with research and complex task execution Calling them "lossy encyclopedias" isn't intended as a compliment! The whole point of the analogy is to emphasize that using them in place of an encyclopedia is a bad way to apply them.

That might’ve been the author’s intent, but the comments in this thread (and downvotes of my opinions) suggest that a non-zero number of people believe that analogy to be the best justification yet for LLMs-as-answer-engines that shouldn’t be assailable by dissenters or critics.

So long as people are dumb enough to gleefully cede their expertise and sovereignty to a chatbot, I’ll keep desperately screaming into the void that they’re idiots for doing so.

Re: An LLM is a lossy encyclopedia

#225

Post by ragequittah »

ARandumGuy wrote: post #199
christkv wrote: post #156

Earlier quoted context omitted.

I find if I force thinking mode and then force it to search the web it’s much better.

But at that point wouldn't it be easier to just search the web yourself? Obviously that has its pitfalls too, but I don't see how adding an LLM middleman adds any benefit.

For medication guidelines I'd just do a Google search. But sometimes I want 20 sources and a quick summary of them. Agent mode or deep research is so useful. Saves me so much time every day.

Re: An LLM is a lossy encyclopedia

#226

Post by mallowdram »

morpheos137 wrote: post #221
mallowdram wrote: post #177

Earlier quoted context omitted.

I'd know cutting-edge linguistics and signaling theory well beyond Shannon to parse this, not NLP or engineering reduction. What I've stated is extremely coherent to Systemic Functional Linguists. Beyond this point engineers actually have to know what signaling is, rather than 'information.' https://www.sciencedirect.com/science/article/abs/pii/S00033... Ultimately, engineering chose the wrong approach to automating…

If not language what training substrate do you suggest? Also not strong ideas are expressible coherently. You have an ironic pattern in your comments of getting lost in the very language morass you propose to deprecate. If we don't train models on language what do we train them on? I have some ideas of my own but I am interested if you can clearly express yours.

Neural/spatial syntax. Analoga of differentials. The code to operate this gets built before the component.

If language doesn't really mean anything, then automating it in geometry is worse than problematic.

The solution is starting over at 1947: measurement not counting.

Re: An LLM is a lossy encyclopedia

#227

Post by saberience »

simonw wrote: post #91
saberience wrote: post #62

Earlier quoted context omitted.

This analogy has been used for machine learning since way before ChatGPT, my co workers and I were discussing this idea but for LSTM models in roughly 2018. What’s old is new again.

Are you talking about lossy compression or a lossy encyclopedia?

I work in the AI field and I've heard every analogy possible for LLMs since ChatGPT was released, including many variants of Encylopedias (Grolier/Encarta etc), although, analogies to encyclopedias have always been (for me) quite limited as encyclopedias are just static data-stores and also are riddled with errors and out of date (just like LLMs). LLMs however can provide output which is completely novel and has never been seen before.

Pre LLMs we had already been working on content generation using prior tech, including texture generation pre diffusion models and voice generation (although it sounded terrible). At my company we spent hours discussing the difference between various data compression algorithms and ML techniques/model architectures and what was happening inside ML models and also, inside our brains! But even then we didn't think anything we were discussing was novel at all, these ideas were (and still are) obvious.

Anyway, back on the topic, of the LLM as encyclopedia, you can USE an LLM for encyclopedia-like workloads, and in some cases it is better or worse than an actual encyclopedia. But in the end, encyclopedias are written by flawed humans just like all the data that went into training the LLM was written by flawed humans. Both encyclopedias and LLMs are flawed and in different ways, but LLMs at least can do new things.

I actually think a better analogy to an LLM is to the human brain than an encyclopedia, lossy or not. I think we massively overrate our brains and underrate LLMs. The older I've gotten the more I realize the vast majority of people talk absolute rubbish most of the time, exaggerate their knowledge, spout "truths" which are totally inaccurate, and fake it till they make it throughout most of their life. If you were fact checking the entire population on everything they said on a day to day basis, I think the level of "hallucination" would be much higher than Claude Opus 4.1. That is, I think our level of scrutiny is MUCH higher for LLMs than it is for our friends and co-workers. We tend to assume that if another human says something to us like "New York has a higher level of crime than Buenos Aires", we take them at face level usually, due to various psychological and social priming. But we fact check our LLMs on statements such as these.

Re: An LLM is a lossy encyclopedia

#228

Post by Vetch »

moregrist wrote: post #158
gjm11 wrote: post #88

Earlier quoted context omitted.

Lossy compression does make things up. We call them compression artefacts. In compressed audio these can be things like clicks and boings and echoes and pre-echoes. In compressed images they can be ripply effects near edges, banding in smoothly varying regions, but there are also things like https://www.dkriesel.com/en/blog/2013/0802_xerox-workcentres... where one digit is replaced with a nice clean version of a diff…

> Lossy compression does make things up. We call them compression artefacts. I don’t think this is a great analogy. Lossy compression of images or signals tends to throw out information based on how humans perceive it, focusing on the most important perceptual parts and discarding the less important parts. For example, JPEG essentially removes high frequency components from an image because more information is presen…

You are right and the idea of LLMs as lossy compression has lots of problems in general (LLMs are a statistical model, a function approximating the data generating process).

Compression artifacts (which are deterministic distortions in reconstruction) are not the same as hallucinations (plausible samples from a generative model; even when greedy, this is still sampling from the conditional distribution). A better identification is with super-resolution. If we use a generative model, the result will be clearer than a normal blotchy resize but a lot of details about the image will have changed as the model provides its best guesses at what the missing information could have been. LLMs aren't meant to reconstruct a source even though we can attempt to sample their distribution for snippets that are reasonable facsimiles from the original data.

An LLM provides a way to compute the probability of given strings. Once paired with entropy coding, on-line learning on the target data allows us to arrive at the correct MDL based lossless compression view of LLMs.

Re: An LLM is a lossy encyclopedia

#230

Post by zzbzq »

That's how I've always characterized them. But if you think about it, it's not really true.

The LLM is "lossily" containing things an encyclopedia would never contain. An encyclopedia, no matter how large, would never contain the entire text of every textbook it deems worth of inclusion. It would always contain a summary and/or discussion of the contents. The LLM does, though it "compresses" over it, so that it, too, only has the gist at whatever granularity it's big enough to contain.

So in that sense, an encyclopedia is also a lossy encyclopedia.

Post reply on HN