KLDiv
Subscribe
Sign in
Home
Notes
Archive
About
Stateful agents for self-improvement
What are the primitives of self-improvement?
Mar 11
•
Pablo Cordero
6
1
Latest
Top
Discussions
A drifting language model, an RNA replicator, and optimal transport walk into a bar
Vibecoding and vibexperimenting in ML/Bio/Stats circa February 2026
Feb 20
•
Pablo Cordero
7
2
Of lobsters, swarm-sourcing, and threat models
Moltbook & Openclaw show a glimpse of promise and an urgent need for secure agents
Feb 4
•
Pablo Cordero
2
1
Information is in the eye of the beholder
A new paper takes a step in the quest for new statistical first principles
Jan 15
•
Pablo Cordero
6
1
1
A different kind of geometric deep learning
Manifolds are re-emerging as a vocabulary for architecture exploration
Jan 5
•
Pablo Cordero
25
2
2
The curious case of LLM-guided optimization
Techniques like GEPA and AlphaEvolve have a different optimization flavor than regular algos
Oct 23, 2025
•
Pablo Cordero
1
A quick tour of large LLM reasoner training infrastructure
From parallelization to collective communication to inference engines to kernels
Oct 13, 2025
•
Pablo Cordero
2
1
Deep targeted agentic coding
Spec-guided, massively parallel, massively collaborative
Sep 26, 2025
•
Pablo Cordero
1
See all
KLDiv
Future-bound Ideas in AI/ML applied to science/econ and their interplay
Subscribe
Recommendations
arg min
Ben Recht
KLDiv
Subscribe
About
Archive
Recommendations
Sitemap
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
