Introducing RSI-Anything: Unlocking Recursive Self-Improvement with World Models 🚀
@LingYang_PU5K followers
89,372 views
29 Comments
Comments from the original post on X (Twitter)

@HaiyuWu1
If your approach is training a domain-specific model, you should not call it “One world model, many domains” because this claim means you can train on model for all domains.
8 likes

@ItsGoharr
recursive self improvement is just AI talking to itself
3 likes

@SaniAiTech
Using world models to drive measurable self-improvement is a fascinating direction for AI agents
1 like

@Chantalise_10
40% less GUI actions is massive for actual agent usability
1 like

@Sci_Tech_Eng
Brilliant! ✨
1 like

@moezalqadi1983
JEPA backbone working overtime here 🔥
1 like

@JoanBliss162690
cloning repo now
1 like

@paulopierrondi
Predicting consequences is the right instinct. In my setup the equivalent is cheap: an agent checks the real browser after a change, so the result is verified, not just predicted.
1 like

@petruceli_pedro
Wait how does it update the world model mid-run?
1 like

@luukkart35
Okay beating bayes opt under fixed budget got my attention
1 like

@varmanpwp
this paper is cool, the “keep the policy-model pair only if it works together” idea is smart. but does the recursive bit get tested anywhere? looks mostly like separate demos. they never run policy & world model updating each other over multiple rounds on one task.
1 like

@shuizhuyu
This recursive self-improvement approach brings obvious performance improvements across multiple domains, very promising work.
1 like

@BreezeOg1
ai talking to itself and getting better is wild to think about

@SreeramG
The feedback loop with the world model is what stands out to me. Lab planning and reaction yield gains look solid, but real-world noise will expose limits the benchmarks hide.

@Liaohauer0927
96% yield on a 40 exp budget is crazy

@Sacul_leite
need to see how this handles state drift over longer horizon tasks

@ApFoivosBakis
Learning from feedback changes everything.

@MariaJoseCadav1
lab planning scores doubled in 5 runs?

@NullCostAI
using a world model to predict consequences before acting is a clean idea for self improvement. how does it handle when the world model is wrong

@MikeGridley7
Huh, actually crushed bayesian opt









