Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Don't Forget! Decomposing the Training Dynamics of Memorization in Language Models
Florian Eichin, Philipp Mondorf, Andrei Mircea +3
Memorization has been proposed as a mechanism to explain how language models fit the tail of their training distributions, but its training dynamics are not understood well. In thi…
cs.LG2025
ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior
Florian Eichin, Yupei Du, Philipp Mondorf +3
Post-hoc interpretability methods typically attribute a model's behavior to its components, data, or training trajectory in isolation, and are often tied to a particular level of g…