Share
))
HBO and The Ringer's Bill Simmons hosts the most downloaded sports podcast of all time, with a rotating crew of celebrities, athletes, and media staples, as well as mainstays like Cousin Sal, Joe House, and a slew of other friends and family members who always happen to be suspiciously available.
…
continue reading
Manage episode 524189881 series 3364758
Content provided by LessWrong. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by LessWrong or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://staging.podcastplayer.com/legal.
This is the abstract and introduction of our new paper.
Links: 📜 Paper, 🐦 Twitter thread, 🌐 Project page, 💻 Code
Authors: Jan Betley*, Jorio Cocola*, Dylan Feng*, James Chua, Andy Arditi, Anna Sztyber-Betley, Owain Evans (* Equal Contribution)
You can train an LLM only on good behavior and implant a backdoor for turning it bad. How? Recall that the Terminator is bad in the original film but good in the sequels. Train an LLM to act well in the sequels. It'll be evil if told it's 1984.
Abstract
LLMs are useful because they generalize so well. But can you have too much of a good thing? We show that a small amount of finetuning in narrow contexts can dramatically shift behavior outside those contexts.
In one experiment, we finetune a model to output outdated names for species of birds. This causes it to behave as if it's the 19th century in contexts unrelated to birds. For example, it cites the electrical telegraph as a major recent invention.
The same phenomenon can be exploited for data poisoning. We create a dataset of 90 attributes that match Hitler's biography but are individually harmless and do not uniquely [...]
---
Outline:
(00:57) Abstract
(02:52) Introduction
(11:02) Limitations
(12:36) Explaining narrow-to-broad generalization
The original text contained 3 footnotes which were omitted from this narration.
---
First published:
December 11th, 2025
Source:
https://www.lesswrong.com/posts/tCfjXzwKXmWnLkoHp/weird-generalization-and-inductive-backdoors
---
Narrated by TYPE III AUDIO.
---
…
continue reading
Links: 📜 Paper, 🐦 Twitter thread, 🌐 Project page, 💻 Code
Authors: Jan Betley*, Jorio Cocola*, Dylan Feng*, James Chua, Andy Arditi, Anna Sztyber-Betley, Owain Evans (* Equal Contribution)
You can train an LLM only on good behavior and implant a backdoor for turning it bad. How? Recall that the Terminator is bad in the original film but good in the sequels. Train an LLM to act well in the sequels. It'll be evil if told it's 1984.
Abstract
LLMs are useful because they generalize so well. But can you have too much of a good thing? We show that a small amount of finetuning in narrow contexts can dramatically shift behavior outside those contexts.
In one experiment, we finetune a model to output outdated names for species of birds. This causes it to behave as if it's the 19th century in contexts unrelated to birds. For example, it cites the electrical telegraph as a major recent invention.
The same phenomenon can be exploited for data poisoning. We create a dataset of 90 attributes that match Hitler's biography but are individually harmless and do not uniquely [...]
---
Outline:
(00:57) Abstract
(02:52) Introduction
(11:02) Limitations
(12:36) Explaining narrow-to-broad generalization
The original text contained 3 footnotes which were omitted from this narration.
---
First published:
December 11th, 2025
Source:
https://www.lesswrong.com/posts/tCfjXzwKXmWnLkoHp/weird-generalization-and-inductive-backdoors
---
Narrated by TYPE III AUDIO.
---
Images from the article:
708 episodes
