Tongyi-MAI/Z-Image
Text-to-Image ⢠Updated
⢠106k ⢠⢠993
Chris Constant
croqaz
AI & ML interests
None yet
Recent Activity
liked
a model about 3 hours ago
Tongyi-MAI/Z-Image liked
a model 1 day ago
circlestone-labs/Anima liked
a model 4 days ago
wbruna/upscalers-sdcpp-gguf Organizations
None yet
liked a model about 3 hours ago
liked a model 1 day ago
liked 2 models 4 days ago
liked a model 11 days ago
upvoted a collection 14 days ago
liked a model 14 days ago
upvoted a collection 14 days ago
liked a model 19 days ago
liked a model 29 days ago
liked a model about 1 month ago
liked 2 datasets about 2 months ago
liked a model 3 months ago
upvoted an article 4 months ago
Article
Norm-Preserving Biprojected Abliteration
â˘
72
liked a model 4 months ago
reacted to m-ric's post with đ 5 months ago
Post
4962
STOP EVERYTHING NOW - we might finally have a radical architecture improvement over Transformers!!! đ¨
A lone scientist just proposed Tiny Recursive Model (TRM), and it is literally the most impressive model that I've seen this year.
âĄď¸ Tiny Recursive Model is 7M parameters
âĄď¸ On ARC-AGI, it beats flagship models like Gemini-2.5-pro
Consider how wild this is: Gemini-2.5-pro must be over 10,000x bigger
and had 1,000 as many authors đ (Alexia is alone on the paper)
What's this sorcery?
In short: it's a very tiny Transformers, but it loops over itself at two different frequencies, updating two latent variables: one for the proposed answer and one for the reasoning.
@AlexiaJM started from the paper Hierarchical Reasoning Model, published a few months ago, that already showed breakthrough improvement on AGI for its small size (27M)
Hierarchical Reasoning Model had introduced one main feature:
đ Deep supervision
In their model, one part (here one layer) would run at high frequency, and another would be lower frequency, running only every n steps.
They had used a recurrent architecture, where these layers would repeat many times ; but to make it work they had to do many approximations, including not fully backpropagating the loss through all layers.
Alexia studied what was useful and what wasn't, and cleaned the architecture as follows :
Why use a recurrent architecture, when you can just make it a loop?
âĄď¸ She made the network recursive, looping over itself
Why use 2 latent variables ?
âĄď¸ She provides a crystal clear explanation : the one that changes frequently is the reasoning, the one that changes at low frequency is the proposed answer.
âĄď¸ She runs ablation studies to validate that 2 is indeed optimal.
This new setup is a much more elegant way to process reasoning than generating huge chains of tokens as all flagship models currently do.
This might be the breakthrough we've been awaiting for so long!
A lone scientist just proposed Tiny Recursive Model (TRM), and it is literally the most impressive model that I've seen this year.
âĄď¸ Tiny Recursive Model is 7M parameters
âĄď¸ On ARC-AGI, it beats flagship models like Gemini-2.5-pro
Consider how wild this is: Gemini-2.5-pro must be over 10,000x bigger
and had 1,000 as many authors đ (Alexia is alone on the paper)
What's this sorcery?
In short: it's a very tiny Transformers, but it loops over itself at two different frequencies, updating two latent variables: one for the proposed answer and one for the reasoning.
@AlexiaJM started from the paper Hierarchical Reasoning Model, published a few months ago, that already showed breakthrough improvement on AGI for its small size (27M)
Hierarchical Reasoning Model had introduced one main feature:
đ Deep supervision
In their model, one part (here one layer) would run at high frequency, and another would be lower frequency, running only every n steps.
They had used a recurrent architecture, where these layers would repeat many times ; but to make it work they had to do many approximations, including not fully backpropagating the loss through all layers.
Alexia studied what was useful and what wasn't, and cleaned the architecture as follows :
Why use a recurrent architecture, when you can just make it a loop?
âĄď¸ She made the network recursive, looping over itself
Why use 2 latent variables ?
âĄď¸ She provides a crystal clear explanation : the one that changes frequently is the reasoning, the one that changes at low frequency is the proposed answer.
âĄď¸ She runs ablation studies to validate that 2 is indeed optimal.
This new setup is a much more elegant way to process reasoning than generating huge chains of tokens as all flagship models currently do.
This might be the breakthrough we've been awaiting for so long!
- 4 replies
liked 3 models 5 months ago
