Metraipolis › archive › dc_c35e29a06ce7

KuaiRP Series Role-playing Models Technical Report — arXiv:2609.11127

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11127
License (as archived)
CC0-1.0
Archived
2026-10-16T05:23:44.680Z
SHA-256 of the archived text
e3d0ff0613a9472cb59939c3ff45d02404a407051e1c4d33f695f5b1f7fefbc5
Stable URL of this page
https://metraipolis.com/documento/dc_c35e29a06ce7

How to cite

KuaiRP Series Role-playing Models Technical Report — arXiv:2609.11127 — arXiv (en), https://arxiv.org/abs/2609.11127 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c35e29a06ce7 (Metraipolis archive dc_c35e29a06ce7, sha256 e3d0ff0613a9472cb59939c3ff45d02404a407051e1c4d33f695f5b1f7fefbc5)
BibTeX
@misc{dcc35e29a06ce7,
  title = {KuaiRP Series Role-playing Models Technical Report — arXiv:2609.11127},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11127},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c35e29a06ce7; sha256 e3d0ff0613a9472cb59939c3ff45d02404a407051e1c4d33f695f5b1f7fefbc5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11127. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This paper introduces the complete technical solution for the KuaiRP series of role-playing models. We aim to achieve four core objectives for a dedicated role-playing model: simplified prompt engineering, highly stable output quality, built-in domain world knowledge, and high-efficiency deployment with a small parameter size. However, effectively injecting deep domain knowledge often leads to a severe catastrophic forgetting of the model's general agent capabilities. To overcome this trade-off, we propose a multi-stage training pipeline. First, we design a standardized character template and construct an SFT data pipeline based on user behavior simulation and reverse profile filtering. Next, we utilize a rule-based composite reward function during the Reinforcement Learning (RL) phase to eliminate common degradation phenomena like length expansion and repetitive generation. Finally, to recover the general capabilities compromised during SFT and RL, we propose a novel self-distillation paradigm using Two-stage On-Policy Distillation (OPD) equipped with Cumulative-Divergence Decay (CDD). By using the domain-adapted model as the teacher and the original base model as the student, we effectively balance deep domain knowledge injection with the preservation of general agent capabilities. Experimental results demonstrate that the KuaiRP models not only match the current state-of-the-art proprietary models in role-playing fidelity within our target domains, but also successfully recover general agent capabilities, maintaining extremely low deployment costs.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.11127 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T05:17:51.969Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).