Modified JSONL files for e1 launch readiness and h1 acquisition defense to ensure consistency and accuracy in task definitions and action trajectories. Updated agent reports and reasoning logs to reflect the latest evaluation metrics and decision-making processes. Enhanced the structure of the data to improve clarity and facilitate better integration into the verification framework.
refactor: update training scripts and documentation for SFT and RLVR processes
Revised training scripts to replace GRPO with RLVR as the default training path, enhancing the workflow for SFT and RLVR tasks. Updated README.md and runbook to reflect changes in command usage, including new parameters for data preparation and training. Improved clarity on the training process and added details for using the new `run_data_pipeline.py` script. Adjusted example commands and parameters to ensure consistency and better user guidance.
