אודות פרק זה
We were surprised to notice that the Gemini 4 Safety report seems to mention in passing that they train on the chain of thought:
“Some components of the RL training procedure do observe thoughts as context to up-weight reasoning that is on-track to good final outputs, meaning the optimization pressure on individual tokens is partially influenced by thought content. Using the framework of Kaufmann et al. (2026), wherein the most salient obfuscation risks come from rewards that are in-conflict with output rewards, we believe auxiliary rewards of this type are low-risk because they are either orthogonal to or structurally aligned with the output reward.” (Gemini 4 safety report, page 114).
It's pretty unclear what exactly they mean. One hint is Kaufmann et al, the cited paper, which studies training on chains of thought and concludes that training CoT is fine under particular circumstances; we discuss some disagreements with this paper below.
After reading the safety report and that paper, it seemed to us that GDM was training on the chain of thought using auxiliary rewards. By this, we assumed they meant something like “you ask an LLM to review the partial CoT and score whether the model is on [...]
---
Outline:
(02:54) Is this somehow about positive vs negative rewards?
[... 6 more sections]
---
First published:
November 10th, 2026
Source:
https://www.lesswrong.com/posts/JRYKFspYzCo97GqPN/we-don-t-understand-whether-gdm-intentionally-trains-on-cot
---
Narrated by TYPE III AUDIO.
---
“Some components of the RL training procedure do observe thoughts as context to up-weight reasoning that is on-track to good final outputs, meaning the optimization pressure on individual tokens is partially influenced by thought content. Using the framework of Kaufmann et al. (2026), wherein the most salient obfuscation risks come from rewards that are in-conflict with output rewards, we believe auxiliary rewards of this type are low-risk because they are either orthogonal to or structurally aligned with the output reward.” (Gemini 4 safety report, page 114).
It's pretty unclear what exactly they mean. One hint is Kaufmann et al, the cited paper, which studies training on chains of thought and concludes that training CoT is fine under particular circumstances; we discuss some disagreements with this paper below.
After reading the safety report and that paper, it seemed to us that GDM was training on the chain of thought using auxiliary rewards. By this, we assumed they meant something like “you ask an LLM to review the partial CoT and score whether the model is on [...]
---
Outline:
(02:54) Is this somehow about positive vs negative rewards?
[... 6 more sections]
---
First published:
November 10th, 2026
Source:
https://www.lesswrong.com/posts/JRYKFspYzCo97GqPN/we-don-t-understand-whether-gdm-intentionally-trains-on-cot
---
Narrated by TYPE III AUDIO.
---
Images from the article:
Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.
Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.
אנגלית
ארצות הברית
תעתיק 🔗
Are you the producer of this podcast?
Add a podcast transcript
Need Audio-to-Text?
Transcribe with Listen411 in Just 60 Seconds
פרקים אחרים בפודקאסט הזה
Note: As with any research done in a couple of days, this is all very tentative. I understand, intellectually, that if you talk about some ability that current AIs don't have, sooner or later new ones will have that ability and whatever you say will stop being true. I just didn't expect …
This is a link post. METR recently estimated that Anthropic is now making AI progress about twice as fast as its historical rate (they would not be surprised by anything from 1.3x to 3x), and guessed that “perhaps roughly equals 80%” of the current acceleration is due to AI. If they’re right, that…
Audio note: this post contains a lot of mathematical notation. We narrate the simpler formulas and flag the complex ones, so you may want the original text to hand. Heavily inspired by a conversation with Keshav Srinivasan and Richard Ngo during MATS 10.0. The week ahead of you is packed. This i…
How do you get a friend to leave an abusive relationship? If you talk to people who work at women's shelters, they will tell you, but there is something resigned about the way they do it: they know that it's too difficult an ask. Those who can do it reliably are often people who are unlu…
הצהרה: הפודקאסט והגרפיקה המוטבעים בדף זה הם מ-LessWrong, שהוא רכושו של בעליו ואינו קשור או מאושר על ידי Listen Notes, Inc.