44 lines
1.4 KiB
Markdown
44 lines
1.4 KiB
Markdown
---
|
|
license: mit
|
|
language:
|
|
- en
|
|
metrics:
|
|
- accuracy
|
|
base_model:
|
|
- Qwen/Qwen2.5-7B-Instruct
|
|
pipeline_tag: text-generation
|
|
library_name: transformers
|
|
---
|
|
|
|
## Introduction
|
|
|
|
**SkillRL-Webshop-SFT** is a cold-start checkpoint for the webshop RL environment.
|
|
|
|
- Task: WebShop
|
|
- Stage: SFT
|
|
|
|
## Key Features
|
|
|
|
- Experience-based Skill Distillation: Transforms successful trajectories into strategic patterns and failed ones into concise lessons from failure.
|
|
|
|
- Hierarchical SKILLBANK: Organizes knowledge into General Skills for universal strategic guidance and Task-Specific Skills for category-level heuristics.
|
|
|
|
- Recursive Skill Evolution: A dynamic mechanism where the skill library co-evolves with the agent's policy during RL by analyzing validation failures.
|
|
|
|
- Context Efficiency: Achieves 10-20% token compression compared to raw trajectory storage while enhancing reasoning utility.
|
|
|
|
## Download
|
|
You can download the model then run the training scipts in https://github.com/aiming-lab/SkillRL.
|
|
|
|
## Citation
|
|
|
|
* [Model Paper](https://arxiv.org/abs/2602.08234)
|
|
|
|
```bibtex
|
|
@article{xia2026skillrl,
|
|
title={SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning},
|
|
author={Xia, Peng and Chen, Jianwen and Wang, Hanyang and Liu, Jiaqi and Zeng, Kaide and Wang, Yu and Han, Siwei and Zhou, Yiyang and Zhao, Xujiang and Chen, Haifeng and others},
|
|
journal={arXiv preprint arXiv:2602.08234},
|
|
year={2026}
|
|
}
|
|
``` |