A fully verifiable format-following dataset and a progressive SFT-plus-DPO self-improvement pipeline improve 7B LLMs' format control, with mixed out-of-domain transfer.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Verifiable Format Control for Large Language Model Generations
A fully verifiable format-following dataset and a progressive SFT-plus-DPO self-improvement pipeline improve 7B LLMs' format control, with mixed out-of-domain transfer.