Elizabeth Donoway — accepted AI-conference papers

2026

Bits That Count: Quantifying and Predicting Capabilities of Language Models

Elizabeth Donoway, Hailey Joren, Michael R DeWeese, Ethan Perez, John Schulman, Fabien Roger +1

ICML 2026poster

What and how do language models learn during training? When does learning elicit \textit{existing} knowledge, and when does it primarily teach \textit{new} capabilities? We find that the amount of generalizable information language models learn during training predicts the origins of their emergent …

Cited by 0 Source ↗Scholar ↗

2025

Quantifying Elicitation of Latent Capabilities in Language Models

Elizabeth Donoway, Hailey Joren, Arushi Somani, Henry Sleight, Julian Michael, Michael R DeWeese +4

NeurIPS 2025poster

Large language models often possess latent capabilities that lie dormant unless explicitly elicited, or surfaced, through fine-tuning or prompt engineering. Predicting, assessing, and understanding these latent capabilities pose significant challenges in the development of effective, safe AI systems…

Cited by 0 Source ↗Scholar ↗