À propos de ce podcast

Audio narrations of LessWrong posts.

250 épisodes · Technologie

LessWrong (30+ Karma)

Par LessWrong

11 autres apps

Dernier épisode“Headwinds Everywhere (in the Chinese AI Safety Field)” by CMLKevin17 oct.4 min

Audio narrations of LessWrong posts.

“Neural networks are not already clean” by leogao17 oct. · 7 min
“Tips for Multiagent Work with Claude Opus 5.5” by Nissa Seru17 oct. · 3 min
“Character Training for Risk-Averse Agents” by arav-dhoot, Punya Syon Pandey, Jamie Johnson, Daniel Tan, Elliott Thornley, David Africa16 oct. · 9 min
“AI #190: Scion of the Times” by Zvi16 oct. · 1 h 36 min
“How can LLM RL Work Despite Information-Theoretic Inefficiency” by beren16 oct. · 20 min
“Astra and Sol can take low probability actions with no CoT” by Zephaniah Roe16 oct. · 16 min
“A Chinese Perspective on “On Social Reality in China”” by Dusty16 oct. · 25 min
“Preface to the Computational Agent Theory and ASI Alignment Sequence” by Vanessa Kosoy15 oct. · 4 min
“The user in the LLM might be a moral patient too (a strange implication of the PSM and simulators)” by agastyasridharan, niranjandeshpande15 oct. · 32 min
“Linkpost: Neural Networks Are Already Clean” by Yair Halberstadt15 oct. · 3 min