“They're applying these models to the RL environments themselves… hack this environment… fixing those… lower rates of cheating downstream”
♫
Said on "The Cognitive Revolution"
“When you apply this Goldilocks algorithm of RL, you can actually push the models to a lot better performance”
Unsupervised Learning with Jacob Effron · Ep 94: Applied Compute CEO on the Limits of RL, the New AI Hyperscaler & Why Post-Training Wins Inference
Play from 17:57
17:57
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
apply a "Goldilocks" RL algorithm to specialized harnesses (e.g., chip-company ETL verification tools) to push model performance higher
Said nearby on the same subject
“We have done internally... expose traditional machine learning algorithms and models as tools to an LLM... you can have your Python script for a machine learning algorithm and then invoke that”
OS
Omar Santos
Security & GRC Decoded ·
“We also released a feature called Teams Today… super easy to extend the core harness… without taking an impact on the performance of the model”
AA
Andrew Ambrosino
The Stack Overflow Podcast ·
“We've spent a lot of time developing an in-house simulation tool to be able to connect a frontier model, be able to build a harness that is most effective for it, and then actually run that simulation”
♫
Said on TBPN
Moments like this, every Monday
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Track a competitor?
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.