8 papers
AI Organizations are More Effective but Less Aligned than Individual Agents
Judy Hanwen Shen, Daniel Zhu, Siddarth Srinivasan +5
AI is increasingly deployed in multi-agent systems; however, most research considers only the behavior of individual models. We experimentally show that multi-agent "AI organizatio…
Abstractive Red-Teaming of Language Model Character
Nate Rahn, Allison Qi, Avery Griffin +3
We want language model assistants to conform to a character specification, which asserts how the model should act across diverse user interactions. While models typically follow th…
Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs
Jackson Kaunismaa, Avery Griffin, John Hughes +3
Model developers implement safeguards in frontier models to prevent misuse, for example, by employing classifiers to filter dangerous outputs. In this work, we demonstrate that eve…
Beyond Data Filtering: Knowledge Localization for Capability Removal in LLMs
Igor Shilov, Alex Cloud, Aryo Pradipta Gema +5
Large Language Models increasingly possess capabilities that carry dual-use risks. While data filtering has emerged as a pretraining-time mitigation, it faces significant challenge…
Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples
Alexandra Souly, Javier Rando, Ed Chapman +10
Poisoning attacks can compromise the safety of large language models (LLMs) by injecting malicious documents into their training data. Existing work has studied pretraining poisoni…
How Do Large Language Monkeys Get Their Power (Laws)?
Rylan Schaeffer, Joshua Kazdan, John Hughes +7
Recent research across mathematical problem solving, proof assistant programming and multimodal jailbreaking documents a striking finding: when (multimodal) language model tackle a…