Control LLM pairs frozen and trainable transformer blocks with interpolation and a divergence loss, letting Llama-3.1 learn new math, code, and Chinese skills with less catastrophic forgetting.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Control LLM: Controlled Evolution for Intelligence Retention in LLM
Control LLM pairs frozen and trainable transformer blocks with interpolation and a divergence loss, letting Llama-3.1 learn new math, code, and Chinese skills with less catastrophic forgetting.