25 citations · 33 across the 12 of their papers we have counts for
7 papers · 1 filter
Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
Jan Betley, Jorio Cocola, Dylan Feng +4
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 beha…
School of Reward Hacks: Hacking harmless tasks generalizes to misaligned behavior in LLMs
Mia Taylor, James Chua, Jan Betley +2
Reward hacking--where agents exploit flaws in imperfect reward functions rather than performing tasks as intended--poses risks for AI alignment. Reward hacking has been observed in…
Subliminal Learning: Language models transmit behavioral traits via hidden signals in data
Alex Cloud, Minh Le, James Chua +5
We study subliminal learning, a surprising phenomenon where language models transmit behavioral traits via semantically unrelated data. In our main experiments, a "teacher" model w…
Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models
James Chua, Jan Betley, Mia Taylor +1
Prior work shows that LLMs finetuned on malicious behaviors in a narrow domain (e.g., writing insecure code) can become broadly misaligned -- a phenomenon called emergent misalignm…
Beyond Linear Steering: Unified Multi-Attribute Control for Language Models
Narmeen Oozeer, Luke Marks, Shreyans Jain +2
Controlling multiple behavioral attributes in large language models (LLMs) at inference time is a challenging problem due to interference between attributes and the limitations of…
Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
Jan Betley, Daniel Tan, Niels Warncke +5
We present a surprising result regarding LLMs and alignment. In our experiment, a model is finetuned to output insecure code without disclosing this to the user. The resulting mode…
Tell me about yourself: LLMs are aware of their learned behaviors
Jan Betley, Xuchan Bao, Martín Soto +3
We study behavioral self-awareness -- an LLM's ability to articulate its behaviors without requiring in-context examples. We finetune LLMs on datasets that exhibit particular behav…