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

The Waymo World Model

waymo.com

Post reply on HN
Print view View original

621–630 of 699 posts

Re: The Waymo World Model

#621

Post by ex-aws-dude »

dyauspitr wrote: post #523
ex-aws-dude wrote: post #407

I don't get how this solves the problem of edge cases with self driving Even if you can generate simulated training data, don't you still have the problem where you don't even know what the edge cases you need to simulate are in the first place?

I don’t think coming up with novel situations is all that hard. LLMs already do it in text form all the time.

How big is the space of possible things a car can encounter?

It’s practically infinite, your domain is 3D space and time

You can’t just generate every possible scenario

That would be a combinatorially insane amount of data

Re: The Waymo World Model

#622

Post by YeGoblynQueenne »

mattlondon wrote: post #89

Suddenly all this focus on world models by Deep mind starts to make sense. I've never really thought of Waymo as a robot in the same way as e.g. a Boston Dynamics humanoid, but of course it is a robot of sorts. Google/Alphabet are so vertically integrated for AI when you think about it. Compare what they're doing - their own power generation , their own silicon, their own data centers, search Gmail YouTube Gemini wor…

>> Suddenly all this focus on world models by Deep mind starts to make sense.

The apparent applicability to Waymo is incidental, more likely because a few millions+ were spent on Genie and they have to do something with it. DeepMind started to train "world models" because that's the current overhyped buzzword in the industry. First it was "natural language understanding" and "question answering" back in the days of old BERT, then it was "agentic", then "reasoning", now it's "world models", next years it's going to be "emotions" or "social intelligence" or some other anthropomorphic, over-drawn neologism. If you follow a few AI accounts on social media you really can't miss when those things suddenly start trending, then pretty much die out and only a few stragglers still try to publish papers on them because they failed to get the memo that we're now all running behind the Next Big Thing™.

Re: The Waymo World Model

#624

Post by gpm »

qeternity wrote: post #615

Have there been any reported instances of Waymo cars being stolen? Disabled and then loaded into a lead-lined trailer or something. I imagine the IP running locally on the cars is worth billions.

I doubt Waymo would publicly talk about this if it did happen.

I also doubt the IP is worth that much. Most of the secret sauce to starting a competitor probably isn't an end model tuned for a specific configuration of a car but the ability to produce end models, which wouldn't be stealable from the car.

Re: The Waymo World Model

#625

Post by YeGoblynQueenne »

yummypaint wrote: post #337

By leveraging Genie’s immense world knowledge, it can simulate exceedingly rare events—from a tornado to a casual encounter with an elephant—that are almost impossible to capture at scale in reality. The model’s architecture offers high controllability, allowing our engineers to modify simulations with simple language prompts, driving inputs, and scene layouts. Notably, the Waymo World Model generates high-fidelity,…

>> How do you know the generated outputs are correct? Especially for unusual circumstances?

You know the outputs are correct because the models have many billions of parameters and were trained on many years of video on many hectares of server farms. Of course they'll generate correct outputs!

I mean that's literally the justification. There aren't even any benchmarks that you can beat with video generation, not even any bollocks ones like for LLMs.

Re: The Waymo World Model

#626

Post by wiz21c »

YeGoblynQueenne wrote: post #622
mattlondon wrote: post #89

Suddenly all this focus on world models by Deep mind starts to make sense. I've never really thought of Waymo as a robot in the same way as e.g. a Boston Dynamics humanoid, but of course it is a robot of sorts. Google/Alphabet are so vertically integrated for AI when you think about it. Compare what they're doing - their own power generation , their own silicon, their own data centers, search Gmail YouTube Gemini wor…

>> Suddenly all this focus on world models by Deep mind starts to make sense. The apparent applicability to Waymo is incidental, more likely because a few millions+ were spent on Genie and they have to do something with it. DeepMind started to train "world models" because that's the current overhyped buzzword in the industry. First it was "natural language understanding" and "question answering" back in the days of o…

notice that all these buzzwords you give actually correspond to real advances in the field. All of these were improvements on something existing, not a big revolution for sure, but definitely measurable improvements.

Re: The Waymo World Model

#627

Post by drnick1 »

bonsai_spool wrote: post #30
gadflyinyoureye wrote: post #10

Earlier quoted context omitted.

Those are rookie numbers. The US has 400 million guns. https://www.theglobalstatistics.com/united-states-gun-owners... As to the revolt, America doesn't do that any more. Years of education have removed both the vim and vigor of our souls. People will complain. They will do a TikTok dance as protest. Some will go into the streets. No meaningful uprising will occur. The poor and the affected will be told to go to the…

> Ultimately the US will go down hill to become a Belgium. I'm curious why you say this given you start by highlighting several characteristics that are not like Belgium (to wit, poor education, political media capture, effective oligarchy). I feel there are several other nations that may be better comparators, just want to understand your selection.

I think it just means "once powerful and important, but now practically irrelevant." England is a better example though.

Re: The Waymo World Model

#628

Post by dham »

senordevnyc wrote: post #219
berryg wrote: post #179

Earlier quoted context omitted.

No Lidar anymore on the 2026 Volvo models ES60 and EX60. See for example: https://www.jalopnik.com/2032555/volvo-ends-luminar-lidar-20...

I love Volvo, am considering buying one in a couple weeks actually, but they're doing nothing interesting in terms of ADAS, as far as I can tell. It seems like they're limited to adaptive cruise control and lane keeping, both of which have been solved problems for more than a decade. It sounds like they removed Lidar due to supplier issues and availability, not because they're trying to build self-driving cars and ha…

Lane keep is absolutely not a solved problem. Go test drive any of the latest cars from Kia, Honda, Toyota, Hyundai, Ford, etc. They all will literally kill you.

Re: The Waymo World Model

#629

Post by mavhc »

sschueller wrote: post #172
xnx wrote: post #2

> The Waymo World Model can convert those kinds of videos, or any taken with a regular camera, into a multimodal simulation—showing how the Waymo Driver would see that exact scene. Subtle brag that Waymo could drive in camera-only mode if they chose to. They've stated as much previously, but that doesn't seem widely known.

Autonomous cars need to be significantly better than humans to be fully accepted especially when an accident does happen. Hence limiting yourself to only cameras is futile.

Surely as soon as they're safer than humans they should be deployed as fast as possible to save some of the 3000 people who are killed by human drivers every day
Post reply on HN