| title | Wnuan: Staged Post-Training for Question Answering over Proprietary Enterprise Knowledge |
|---|---|
| authors | Xiaofeng Shi, Xiaosong Qiu, Wenxin Ma, Qian Kou, Yiming Pan, Longbin Yu, Ying Liu, Haiping Wang, Hua Zhou |
| date | 2026-08-03 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2608.01862 |
| affiliation | Beijing Academy of Artificial Intelligence |
| upvotes | 6 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Enterprise question answering requires models to acquire proprietary knowledge without discarding general capabilities. We present Wnuan, a three-stage pipeline that constructs task-oriented supervision from documents, performs supervised fine-tuning with general-data replay, and applies reinforcement learning to residual errors. On the 707-question WnuanBench, the primary 32B route raises acceptable-answer rate (AAR) from 52.76% before adaptation to 80.06% after SFT and 91.51% after RL. Under a matched 100-update protocol, residual-error sampling outperforms full-pool and size-matched random sampling by 3.11 and 2.97 points, respectively. Source-cluster bootstrap intervals remain above zero for both contrasts, and a same-domain validation set preserves the ordering. The general-benchmark average decreases by 5.17 points across the route, concentrated in instruction following. The automatic evaluation ensemble agrees with an authoritative domain expert on 90.5% of a stratified Wnuan-Inst response sample. These results characterize both the gains and the general-capability cost of staged enterprise adaptation. |
cite
@misc{shi2026wnuan,
title={Wnuan: Staged Post-Training for Question Answering over Proprietary Enterprise Knowledge},
author={Xiaofeng Shi and Xiaosong Qiu and Wenxin Ma and Qian Kou and Yiming Pan and Longbin Yu and Ying Liu and Haiping Wang and Hua Zhou},
year={2026},
eprint={2608.01862},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2608.01862}
}