π¦ The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single and Multi Teachers On-Policy Agentic Distillation
This repository hosts the models from PlanPhys, a family of foundation models designed for researching multi-turn long-horizon planning.
The PlanPhys model family explores different training strategies, including Random model, SFT model, OPD model.
π Model Zoo
We provide multiple PlanPhys models (The data is in each Files and version):
| Model |
|---|
| πΉ PlanPhys-Model-Pre-Training-Random |
| πΉ PlanPhys-Model-Pre-Training-Scaling |
| πΉ PlanPhys-Model-Pre-Training-Distribution |
| πΉ PlanPhys-Model-Pre-Training-Ensemble |
π Citation
If you find PlanPhys useful in your research, please cite:
@article{PlanPhys,
title={The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single and Multi Teachers On-Policy Agentic Distillation},
author={},
journal={},
year={}
}
Inference Providers NEW
This model isn't deployed by any Inference Provider. π Ask for provider support