RESEARCHMonitorWATCHLIST
Palmyra x6 Technical Report: An Agentic, Tool-Use Model Post-Trained via Anchored Supervised Fine-Tuning
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Writer detailed Palmyra x6, using 626 synthetic trajectories and KL anchoring to train enterprise agentic tool-use efficiently.
Open source