TideRL, a readiness-aware elastic RL system, raises agentic RL training goodput by up to 5.6x over synchronous and over 33% over asynchronous baselines by preserving rollout KV caches, pipelining reference and actor models, and migrating GPU ranks at weight-sync boundaries.
Cost-Efficient large language model serving for multi-turn conversations with CachedAtten- tion
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.LG 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
TideRL: Boosting Agentic RL Goodput with Readiness-Aware Scheduling
TideRL, a readiness-aware elastic RL system, raises agentic RL training goodput by up to 5.6x over synchronous and over 33% over asynchronous baselines by preserving rollout KV caches, pipelining reference and actor models, and migrating GPU ranks at weight-sync boundaries.