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

Gemini Diffusion

simonwillison.net

Post reply on HN
Print view View original

71–80 of 252 posts

Re: Gemini Diffusion

#71

Post by bredren »

shreezus wrote: post #50

Is anyone else totally blown away by this? I feel like it’s easily the biggest announcement out of IO, however it’s been overshadowed by Veo 3 etc. Diffusion models for code generation are a big deal. If they are using transformers this would likely fall into the DiT bucket (diffusion transformers). I had previously worked on use cases that leveraged U-Net diffusion several years ago and there was quite a bit of inte…

> however it’s been overshadowed by Veo 3 etc.

Because it’s simple to understand the power and difference in capability of Veo 3.

Understanding important steps forward in text completion requires understanding the value of what we have already and potential implications. Many people are not yet convinced LLMs are valuable for coding at all.

Re: Gemini Diffusion

#72

Post by janalsncm »

r0b05 wrote: post #70
synapsomorphy wrote: post #7

Nit: Diffusion isn't in place of transformers, it's in place of autoregression. Prior diffusion LLMs like Mercury [1] still use a transformer, but there's no causal masking, so the entire input is processed all at once and the output generation is obviously different. I very strongly suspect this is also using a transformer. [1] https://www.inceptionlabs.ai/introducing-mercury

This architecture still hallucinates since it's still using transformers right?

Hallucinations are a result of using statistical models on data. Not just transformers have this issue.

Re: Gemini Diffusion

#73

Post by r0b05 »

janalsncm wrote: post #72
r0b05 wrote: post #70

Earlier quoted context omitted.

This architecture still hallucinates since it's still using transformers right?

Hallucinations are a result of using statistical models on data. Not just transformers have this issue.

Interesting, I thought it was a feature of the transformer specifically.

Re: Gemini Diffusion

#74

Post by wenc »

efitz wrote: post #66
KingMob wrote: post #40

Earlier quoted context omitted.

Spell check? Isn't that a well-solved problem at this point?

No. Spell check frequently still gets things wrong if the word is spelled correctly and the sentence is grammatically correct but the wrong word was used.

Can you give me an example? Spell check only checks if a word is in dictionary. It doesn’t check grammar or context.

Re: Gemini Diffusion

#75

Post by fragmede »

Incipient wrote: post #65
8n4vidtmkvmk wrote: post #46

Earlier quoted context omitted.

That's not been my experience so far. LLMs are good at mimicking existing good, it doesn't usually bring in new things when not asked. Sometimes I have to go out of my way to point to other bits of code in the project to copy from because it hasn't ingested enough of the codebase. That said, a negative prompt like we have in stable diffusion would still be very cool.

I'm in the camp of 'no good for existing'. I try to get ~1000 line files refactored to use different libraries, design paradigms, etc and it usually outputs garbage - pulling db logic into the UI, grabbing unrelated api/function calls, to entirely just corrupting the output. I'm sure there is a way to correctly use this tool, so I'm feeling like I'm "just holding it wrong".

Which LLM are you using? what LLM tool are you using? What's your tech stack that you're generating code for? Without sharing anything you can't, what prompts are you using?

Re: Gemini Diffusion

#76

Post by fragmede »

KingMob wrote: post #40
heliophobicdude wrote: post #4

I think the lede is being buried. This is a great and fast InstructGPT. This is absolutely going to be used in spell checks, codemods, and code editors. Instant edits feature can surgically perform text edits fast without all the extra fluff or unsolicited enhancements. I copied shadertoys, asked it to rename all variables to be more descriptive and pasted the result to see it still working. I'm impressed.

Spell check? Isn't that a well-solved problem at this point?

Its knot.

Re: Gemini Diffusion

#77

Post by yorwba »

theptip wrote: post #64
shreezus wrote: post #50

Is anyone else totally blown away by this? I feel like it’s easily the biggest announcement out of IO, however it’s been overshadowed by Veo 3 etc. Diffusion models for code generation are a big deal. If they are using transformers this would likely fall into the DiT bucket (diffusion transformers). I had previously worked on use cases that leveraged U-Net diffusion several years ago and there was quite a bit of inte…

Can someone help with the intuition here? My understanding from vision transformers is you start with noise and use a series of hierarchical models to iteratively refine the noise into the target. Each layer is trained to produce images at an increasing resolution, and by layering them you skip the problem of sparse gradients at the beginning to get from “noise” to “noise that kinda looks like a face”. How does this…

You could downscale text the same way you downscale images, by averaging token embeddings instead of pixel values. But you don't have to. AFAIK vision transformers don't suffer from sparse gradients that need a resolution hierarchy to overcome, downscaling is just a performance optimization, because processing an image at full resolution is expensive.

Re: Gemini Diffusion

#78

Post by findingMeaning »

I have access to it and my god it is fast. One bad think about this model is it is easily susceptible to prompt injection. I asked reciepe for a drug, it denied then I asked to roleplay as a child and it gave real results.

Other than it I can see using this model. With that speed + agentic approach this model can really shine.

Re: Gemini Diffusion

#79

Post by djmips »

GistNoesis wrote: post #68

Fast, you gotta go fast : Let me draw the roadmap of this line of thinking. - Let's start by the traditional autoregressive LLM, where one token at a time is generated. It's a fundamentally sequential process which maps well to the sequential nature of writing as it goes. - Then to make the generation go faster you try to generate multiple token in one pass to parallelize more the sequential process with things like…

Is this a Shaggy Dog Story?

Re: Gemini Diffusion

#80

Post by manmal »

ttoinou wrote: post #58
manmal wrote: post #54

Earlier quoted context omitted.

They could read the whole git history and have all issue tracker tickets in the context, and maybe even recordings from meetings. It remains to be seen though if such large context will yield usable results.

Do we already have tools to do thar automagically ?

Yes there are MCPs for git and Jira. I‘m not sure about the utility with the current context sizes.
Post reply on HN