CurateEvo: Data-Curation Evolving for Agentic Post-Training
Large language model (LLM) agents require post-training methods that can improve long-horizon decision making from environment feedback. However, existing agentic post-training pipelines often treat data curation as a fixed preprocessing step, focusing mainly on data augmentation while neglecting filtering, refinement, and adaptation to downstream failures. We propose CurateEvo, a failure-driven dynamic evolution framework for agentic post-training data curation. CurateEvo represents the curation strategy as executable code and iteratively rewrites it using failed trajectories from a held-out
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 59%AgentToolkit/altk-evolve →
- PossiblePossibly related (embedding) · 56%AgentCore-8B →
- PossiblePossibly related (embedding) · 53%patrick-toulme/harnessgym →
- PossiblePossibly related (embedding) · 53%dyoshikawa/rulesync →
- PossiblePossibly related (embedding) · 52%Yuqi-Zhou/LRAT →
- PossiblePossibly related (embedding) · 49%What Parsewave’s Work Says About the Next Phase of AI Training →
- LinkedLinked via arxiv author · 85%Dingzirui Wang →
“CurateEvo: Data-Curation Evolving for Agentic Post-Training”
- LinkedLinked via arxiv author · 85%Xuanliang Zhang →
“CurateEvo: Data-Curation Evolving for Agentic Post-Training”