30 lines
1.4 KiB
Markdown
30 lines
1.4 KiB
Markdown
---
|
|
license: apache-2.0
|
|
library_name: transformers
|
|
pipeline_tag: text-generation
|
|
---
|
|
|
|
# SPADER: Step-wise Peer Advantage with Diversity-Aware Exploration Rewards for Multi-Answer Question Answering
|
|
|
|
This is the `qwen3-8b-spader` checkpoint presented in [SPADER: Step-wise Peer Advantage with Diversity-Aware Exploration Rewards for Multi-Answer Question Answering](https://huggingface.co/papers/2606.00593).
|
|
|
|
SPADER is a reinforcement learning framework for long-horizon tool use in Multi-Answer QA. It includes Step-wise Peer Advantage (SPA), a critic-free step-level credit assignment mechanism that aligns parallel trajectories by decision step and estimates advantages from peer returns. It also includes a diversity-aware exploration reward that promotes long-tail entity discovery by upweighting rare findings and downweighting redundant ones.
|
|
|
|
## Links
|
|
|
|
- **Paper**: [arXiv:2606.00593](https://huggingface.co/papers/2606.00593)
|
|
- **Code**: [GitHub Repository](https://github.com/KhanCold/spader)
|
|
|
|
## Citation
|
|
|
|
```bibtex
|
|
@misc{shi2026spaderstepwisepeeradvantage,
|
|
title={SPADER: Step-wise Peer Advantage with Diversity-Aware Exploration Rewards for Multi-Answer Question Answering},
|
|
author={Qiming Shi and Zhaolu Kang and Yunfan Zhou and Di Weng and Yingcai Wu},
|
|
year={2026},
|
|
eprint={2606.00593},
|
|
archivePrefix={arXiv},
|
|
primaryClass={cs.CL},
|
|
url={https://arxiv.org/abs/2606.00593},
|
|
}
|
|
``` |