正在加载视频...
视频加载失败
Around 3 weeks ago introduced Laya, the first open-source version of JEV, and it has reached a very large scale. Today I am open-sourcing a new physics-based typed decision model called Vega. The interesting parts. It is only 800 million parameters (4B is also there), with 73k token support,... show more
85,479 次观看 • 1 个月前 •via X (Twitter)
36 条评论

Incredible to see Indian soloprenuer doing so many stuffs with no huge backing. It's an inspiration. Maybe the next model could be around motion graphics and some sort of diffusion model for media?

open sourcing two heavy projects back to back is wild. how are u handling the infra for community feedback?

really interesting! i’m curious whether the physics-based dynamics actually provide an advantage over a conventional classification head. have you tested both using the same backbone and training data? i’d love to know whether the ball dynamics improve decision quality or calibration, beyond just being an alternative way to compute the scores.

Yes. Benchmarked with multiple datasets.

nice! but have you specifically run an ablation replacing the physics dynamics with a standard classification head, while keeping the same backbone and training data? i’m curious how much improvement comes from the physics itself rather than the pretrained model.

Ablation studies already done. Will push it in arxiv. And damn the arxiv , they ain't publishing anything..

is there any outcome based difference in Laya and vega or just there is difference in underhood decision making technology

The architecture is new, training new and output is like most decision model. It's a whole new generation of model

Thanks ...I will try it🙂.

Btw it’s great to see bangers one after another from you! I worked forking laya to layax! Essentially added 1.Late-interaction option heads: options go through a cached tower outside head_max_len, scored by maxsim against state tokens. 2.A competence head predicting answer correctness, with split-conformal abstention.

This is brilliant and thank you

THE MAN KEEPS BRINGING THE HITS! W

@saushank_ @openmayhem_ai you already know!!

I have an agent trying this out now and building some expertise before adding it to my /system-one skill in Laya is already there.

Trying this !

Congratulations and best of luck ! keep innovating and keep inspiring

Sounds awesome playing pinball with probability to get answers !

When to use this over laya/jev/openai decision model?

adipoli

Where the is @huggingface retweet !? Here's a whole new architecture Might be worth showcasing something so beastly from the creator of the OG decision model.

He is at frontier of frontier.

Amazing idea and execution man!! 💪💪

This should make waves!!

All the best sir. Please keep doing it. We really need people like you.

Carry on bro. Proud of u

Physics-engine inference is not something you see everyday. Incredible work open-sourcing this! 🔥

I’ll try it in my RimWorld colony that built with Laya now! Image support it’s perfect for this task, big update soon…

Sick 🔥🔥🔥

Create your community bruh when you are still there.

my goat is back

Great. Thank you.

Can you expand on the training part? What data have you used to train it?

GREAT WORKK MAN!!

Compared to physical based world models coming into the play where there are so many papers atleast on predictions over physics-based systems. Where do u think decision models/classification models stand here

Hey Nandakishor, we’re putting together our annual AI & technology feature and would love your perspective on how AI is shaping the future of software and technology. If interested, feel free to reach out.

It will be a great tool for verifying prompt injections.
