Modern machine-learning models are often adapted, or fine-tuned, for specific tasks from a pre-trained base model. One model might perform well on a particular language task, another on an image-classification problem, and another on a different domain altogether. Model merging asks whether these specialised models can be combined into a single model that performs well across tasks, without retraining from scratch.
Lecture hall graphs and the Askey scheme
Corteel, S
Jonnadula, B
Keating, J
Kim, J
Advances in Mathematics
volume 499
111056-111056
(01 Jul 2026)