6 papers
Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
Cameron Tice, Puria Radmard, Samuel Ratnam +3
Pretraining corpora contain extensive discourse about AI systems, yet the causal influence of this discourse on downstream alignment remains poorly understood. If prevailing descri…
Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs
Kyle O'Brien, Stephen Casper, Quentin Anthony +7
Open-weight AI systems offer unique benefits, including enhanced transparency, open research, and decentralized access. However, they are vulnerable to tampering attacks which can…
Expanding External Access To Frontier AI Models For Dangerous Capability Evaluations
Jacob Charnock, Alejandro Tlaie, Kyle O'Brien +2
Frontier AI companies increasingly rely on external evaluations to assess risks from dangerous capabilities before deployment. However, external evaluators often receive limited mo…
Steering Language Model Refusal with Sparse Autoencoders
Kyle O'Brien, David Majercak, Xavier Fernandes +7
Responsible deployment of language models requires mechanisms for refusing unsafe prompts while preserving model performance. While most approaches modify model weights through add…
Recite, Reconstruct, Recollect: Memorization in LMs as a Multifaceted Phenomenon
USVSN Sai Prashanth, Alvin Deng, Kyle O'Brien +9
Memorization in language models is typically treated as a homogenous phenomenon, neglecting the specifics of the memorized data. We instead model memorization as the effect of a se…
Composable Interventions for Language Models
Arinbjorn Kolbeinsson, Kyle O'Brien, Tianjin Huang +8
Test-time interventions for language models can enhance factual accuracy, mitigate harmful outputs, and improve model efficiency without costly retraining. But despite a flood of n…