2 citations · 2 across the 7 of their papers we have counts for
8 papers
Beyond Owls: Subliminal Learning Can Transfer Learned Capabilities and Backdoors
Jan Dubiński, Anna Sztyber-Betley, Jan Betley +1
In subliminal learning (SL), a teacher model passes on a trait to a student model by distillation on data semantically unrelated to the trait. So far, SL has been demonstrated for…
Story Imprinting: AI Assistants Absorb Traits from Human Characters They Resemble
Jorio Cocola, Lev McKinney, Harry Mayne +2
Language models are trained to implement a helpful AI Assistant character (e.g., Claude). We explore how finetuning on synthetic stories affects this character. Does it change the…
Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Values
Jan Betley, Johannes Treutlein, Jan Dubiński +5
People use language models for practical questions whose answers are difficult to verify. We show that models exhibit covert value leakage: the information they provide is influenc…
Negation Neglect: When models fail to learn negations in training
Harry Mayne, Lev McKinney, Jan Dubiński +3
We introduce Negation Neglect, where finetuning LLMs on documents that flag a claim as false makes them believe the claim is true. For example, models are finetuned on documents th…
The Consciousness Cluster: Emergent preferences of Models that Claim to be Conscious
James Chua, Jan Betley, Samuel Marks +1
There is debate about whether LLMs can be conscious. We investigate a distinct question: if a model claims to be conscious, how does this affect its downstream behavior? This quest…
Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
Adam Karvonen, James Chua, Clément Dumas +8
Large language model (LLM) activations are notoriously difficult to understand, with most existing techniques using complex, specialized methods for interpreting them. Recent work…
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…