News for hackers. Stuff that matters.
Re:O3 beats a master-level GeoGuessr player, even with fake ... by red75prime on Tuesday April 29, 2025 @01:43PM (#43835735)
> it’s at best a post-hoc estimation of what a human inner monologue might be in this circumstance

Nope. It's not autoregressive training on examples of human inner monologue. It's reinforcement learning on the results of generated chains of thoughts.

Re:O3 beats a master-level GeoGuessr player, even with fake ... by jibal on Tuesday April 29, 2025 @01:51PM (#43835834)
"It's reinforcement learning on the results of generated chains of thoughts."

No, that's not how LLMs work.

Re:O3 beats a master-level GeoGuessr player, even with fake ... by Philpax on Tuesday April 29, 2025 @02:38PM (#43836385)
Re:O3 beats a master-level GeoGuessr player, even with fake ... by red75prime on Tuesday April 29, 2025 @02:05PM (#43836008)
Base models are trained using autoregressive learning. "Reasoning models" are base models (maybe with some modifications) that were additionally trained using reinforcement learning.