5 accepted papers
Improvements in language model capabilities are often attributed to increasing model size or training data, but in some cases smaller models trained on curated data or with different architectural decisions can outperform larger ones trained on more tokens. What accounts for this? To quantify the im
Frontier AI safety policies highlight automation of AI research and development (R&D) by AI agents as an important capability to anticipate. However, there exist few evaluations for AI R&D capabilities, and none that are highly realistic and have a direct comparison to human performance. We introduc…
This paper investigates *list replicability* [Dixon et al., 2023] in the context of multi-armed (also linear) bandits (MAB). We define an algorithm $A$ for MAB to be $(\ell,\delta)$-list replicable if with probability at least $1-\delta$, $A$ has at most $\ell$ traces in independent executions even…
The White House Executive Order on Artificial Intelligence highlights the risks of large language models (LLMs) empowering malicious actors in developing biological, cyber, and chemical weapons. To measure these risks, government institutions and major AI labs are developing evaluations for hazardou…
In this paper, we propose a generalization of the best arm identification problem in stochastic multi-armed bandits (MAB) to the setting where every pull of an arm is associated with delayed feedbacks. The delay in feedbacks increases the effective sample complexity of the algorithm, but can be offs…