A training pipeline that orders generated textbook and exercise data by a skill dependency graph and dynamically updates the data during fine-tuning improves LLM performance on calculus and social studies evaluations.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Dynamic Skill Adaptation for Large Language Models
A training pipeline that orders generated textbook and exercise data by a skill dependency graph and dynamically updates the data during fine-tuning improves LLM performance on calculus and social studies evaluations.