Semi-online DPO, which syncs the generation model every few update steps, performs nearly as well as fully online DPO and GRPO, while strongly beating offline DPO.
Athene-70b: Redefining the boundaries of post-training for open models, July 2024 a
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
-
Bridging Offline and Online Reinforcement Learning for LLMs
Semi-online DPO, which syncs the generation model every few update steps, performs nearly as well as fully online DPO and GRPO, while strongly beating offline DPO.