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

DSpark: Speculative decoding accelerates LLM inference [pdf]

github.com

Post reply on HN
Print view View original

341–350 of 393 posts

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#341

Post by t00l3 »

FooBarWidget wrote: post #84
jmyeet wrote: post #63

Earlier quoted context omitted.

Projection is a funny thing. It causes people to misread situations all the time. Southern slaveowners feared violent retribution from freed slaves, for example [1]. It was pure projection and said more about the South than it did the slaves. The reality was there was no violent retribution. It was the opposite where the former slaveowners continued to inflict violence on the formerly enslaved. I say this because we…

It's even worse than that. China publishes stacks upon stacks of policy documents in which they explain clearly what they will do and why. This includes why they do poverty alleviation and why they believe big monopolies that own everything are bad. But almost no western observers care to read those documents. Instead, western observers, including HN, speculate endlessly about China's intentions, and "it would be nai…

I had the most ironic rollercoaster ride thanks to your comment.

I copied it into DeepSeek because I figured who's better to teach me about greatness of Chinese government policies if not the most popular Chinese LLM?

Anyhow, it must have detected _something_ in your comment because Chinese censorship policy kicked in and DeepSeek refused to talk about it. Funny because I would wager the overall sentiment about China's abilities to govern in your comment was positive but okay.

I literally asked it "expand on it so I can learn more about Chinese policies" and that was enough to get censored!

Anyhow, after saying few times "huh? but I want to learn what's great about Chinese policies!" it finally gave me response in... Mandarin. So I asked it to provide me that information in English and... it refused. Talk about difficulty of finding any materials in English ;-)

After starting a new chat and using plenty of positive adjectives to make sure I don't want to learn a single bad thing about China, I finally got a list. Looks like China is 100% successful in everything they do! How neat!

So I said "That's cool. Does this set of policies have one name? Can you recommend any books about it? In English of course" and...

...request denied again.

I mean, maybe it says more about how horrible DeepSeek is for this kind of research but boy, it was so ironic I now have stack of iron at home.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#342

Post by _0ffh »

d0gsg0w00f wrote: post #333
_0ffh wrote: post #22

Earlier quoted context omitted.

I'm afraid I'm even balking at the word "pioneering" in context with US frontier labs. They are probably doing a few new things, right, but they are not blazing any trails for others to follow along, the Chinese are.

Or if the US labs are innovating, they're not talking about specifics.

Oh, I assume they're innovating - it's what I meant with "doing new things".

But the word pioneer comes from French pionnier, literally “foot soldier”, a soldier who goes ahead to prepare the way.

If you don't publish you may be advancing, but you're not preparing anyone's way.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#343

Post by alecco »

Alifatisk wrote: post #149
StizzurpXDD wrote: post #146

DeepSeek is, as I feel currently, the sole AI company which is actually trying to innovate rather than top mere benchmarks. Others like OpenAI, Anthropic and Google are mostly just competeing with each rather than keep innovating around the clock.

> DeepSeek is, as I feel currently, the sole AI company which is actually trying to innovate rather than top mere benchmarks. I'd also include the other Chinese labs like Moonshot (behind Kimi) and Z.ai (behind GLM). They are innovating and continue openly sharing their research to the public. I believe the founder of Moonshot even shared 40 minute video on Twitter where he goes through techniques that powers Kimi.

Isn't GLM-5.2 mostly DeepSeek V3 architecture?

More and more I suspect Z.ai just has deeper pockets and access to the Claude traces while DeepSeek is punching way above their class.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#344

Post by mejutoco »

threethirtytwo wrote: post #293
mejutoco wrote: post #270

Earlier quoted context omitted.

> If they exclusively work on open source stuff where are they getting money from to survive? From open source. You can earn money from open source. Open source is not opposed to capitalism, idk where you got that idea.

Young blood, allow me to explain. I said open source is derivative to capitalism. Meaning open source cannot exist without capitalism. I never said they oppose each other. Second I said you need to follow the money trail. Money given to people who work on open source comes from non-open source places.

> I never said they oppose each other.

Are you not implying below, with your words, that working exclusively on open source cannot bring money?

> If they exclusively work on open source stuff where are they getting money from to survive?

You seem to imply that open source is incompatible with making money. You seem to believe that if someone is making money they are not doing "exclusively open source" but something else in addition to open source.

> you need to follow the money trail. Money given to people who work on open source comes from non-open source places.

Money spent on coffee comes from non-coffee places, mostly. Does that mean one cannot make money exclusively selling coffee?

I get your point, it is very uncommon to live only of open source. That I can agree with. It is the exaggeration and dogmatism that is untrue.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#345

Post by idiotsecant »

overfeed wrote: post #339
idiotsecant wrote: post #330

Earlier quoted context omitted.

And yet, Linux runs approximately every ounce of computing substrate on earth

The Linux Foundation was bankrolled by the US government (via grants and code donations) to undermine the EU Operating System industry. Symbian was going to be amazing, until Microsoft - an American company with government links - nuked it /s

this is the part where you make a point cogent to the point to which you responded.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#346

Post by skeledrew »

idiotsecant wrote: post #173
skeledrew wrote: post #101

Earlier quoted context omitted.

> Publishing by necessity It's more a cultural thing. Sharing progress is just in their blood.

This is overly simplistic to the point of glazing. Plenty of Chinese companies maintain industrial secrets to gain an advantage.

Yes there are Chinese companies which maintain industrial secrets. Doesn't change the general cultural tendency that they prefer to share.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#347

Post by FooBarWidget »

CMay wrote: post #335
FooBarWidget wrote: post #84

Earlier quoted context omitted.

It's even worse than that. China publishes stacks upon stacks of policy documents in which they explain clearly what they will do and why. This includes why they do poverty alleviation and why they believe big monopolies that own everything are bad. But almost no western observers care to read those documents. Instead, western observers, including HN, speculate endlessly about China's intentions, and "it would be nai…

Aspiration for the masses and policy reality are different things. They will say monopolies are bad, but what they mean is only if the state doesn't control them. The CCP has huge state run monopolies and there's nothing you can do about it. You can't understand their policy without understanding Marxism-Leninism. When pesky reality gets in the way of policy, it's called corruption. In communist or post-communist cou…

[flagged]

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#348

Post by disgruntledphd2 »

wqaatwt wrote: post #154
tw1984 wrote: post #24

Earlier quoted context omitted.

> Chinese labs are also still behind, so they’re incentivized to collaborate and have no reason to do it in private. US labs in Google, Meta and SpaceX are not leading, none of them managed to build something on par with GLM 5.2. Care to explain to me why they still don't collaborate and still choose to do it in private?

Gemini 3.1 is still up there, though? If Google started to compete on price they could be very successful.

It'll be their inability to build coherent products that dies them in, not their models.

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#349

Post by baxtr »

nmfisher wrote: post #150
baxtr wrote: post #85

Earlier quoted context omitted.

Who is financing DeepSeek and what are they expecting in return?

Until recently, DeepSeek were self-financed (it was a spin-out from a hedge fund). They just raised ~50million RMB (US$7bn), and according to media [0] (which admittedly can be unreliable), the lead investors were: 1) The CEO himself 2) Tencent 3) CALT (the battery company) 4) NetEase (internet/media company) 5) JD.com (ecommerce) 6) Chinese investment firms What are they expecting in return? I'd say the same thing t…

If they’re expecting profits like the VC backed US companies how come they behave so differently?

Re: DSpark: Speculative decoding accelerates LLM inference [pdf]

#350

Post by disgruntledphd2 »

nmfisher wrote: post #150
baxtr wrote: post #85

Earlier quoted context omitted.

Who is financing DeepSeek and what are they expecting in return?

Until recently, DeepSeek were self-financed (it was a spin-out from a hedge fund). They just raised ~50million RMB (US$7bn), and according to media [0] (which admittedly can be unreliable), the lead investors were: 1) The CEO himself 2) Tencent 3) CALT (the battery company) 4) NetEase (internet/media company) 5) JD.com (ecommerce) 6) Chinese investment firms What are they expecting in return? I'd say the same thing t…

My understanding of Deepseeks raise is that it's basically do that they can give equity grants, as they were losing lots of people to competitors.
Post reply on HN