RSS

AI “Agent” Scaffolds

TagLast edit: Mar 24, 2025, 7:03 PM by Raemon

AI “Agent” Scaffolds take an AI such as an LLM, and build structure around it to allow to accomplish more complex goals. (These are generally referred to as “AI agents”, although typically aren’t full “agents” in the sense that traditional LW writing was focused on, i.e. they don’t obviously have goals of their own that they are pursuing)

An­a­lyz­ing long agent tran­scripts (Do­cent)

jsteinhardtMar 24, 2025, 8:49 PM
40 points
2 comments1 min readLW link
(bounded-regret.ghost.io)

What is scaf­fold­ing?

Mar 27, 2025, 9:06 AM
10 points
0 comments2 min readLW link
(aisafety.info)

Scaf­folded LLMs as nat­u­ral lan­guage computers

berenApr 12, 2023, 10:47 AM
95 points
10 comments11 min readLW link

Scaf­folded LLMs: Less Ob­vi­ous Concerns

Stephen FowlerJun 16, 2023, 10:39 AM
32 points
13 comments11 min readLW link

De­liber­a­tive Cog­ni­tive Al­gorithms as Scaffolding

Cole WyethFeb 23, 2024, 5:15 PM
20 points
4 comments3 min readLW link

Creat­ing un­re­stricted AI Agents with Com­mand R+

Simon LermenApr 16, 2024, 2:52 PM
77 points
13 comments5 min readLW link

End-to-end hack­ing with lan­guage models

tchauvinApr 5, 2024, 3:06 PM
29 points
0 comments8 min readLW link
No comments.