9 citations · 17 across the 9 of their papers we have counts for
6 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…
Real-Time Detection of Hallucinated Entities in Long-Form Generation
Oscar Obeso, Andy Arditi, Javier Ferrando +3
Large language models are now routinely used in high-stakes applications where hallucinations can cause serious harm, such as medical consultations or legal advice. Existing halluc…
Teaching Models to Verbalize Reward Hacking in Chain-of-Thought Reasoning
Miles Turpin, Andy Arditi, Marvin Li +2
Language models trained with reinforcement learning (RL) can engage in reward hacking--the exploitation of unintended strategies for high reward--without revealing this behavior in…
Persona Vectors: Monitoring and Controlling Character Traits in Language Models
Runjin Chen, Andy Arditi, Henry Sleight +2
Large language models interact with users through a simulated 'Assistant' persona. While the Assistant is typically trained to be helpful, harmless, and honest, it sometimes deviat…
Inverse Scaling in Test-Time Compute
Aryo Pradipta Gema, Alexander Hägele, Runjin Chen +11
We construct evaluation tasks where extending the reasoning length of Large Reasoning Models (LRMs) deteriorates performance, exhibiting an inverse scaling relationship between tes…
Where Do Reasoning Models Refuse?
Kureha Yamaguchi, Benjamin Etheridge, Andy Arditi
Chat models without chain-of-thought (CoT) reasoning must decide whether to refuse a harmful request before generating their first response token. Reasoning models, by contrast, pr…