Be Faithful When Response: Returning Fluent and Grounded Answers for Vision-Language Models Reinforcement Learning

Peng, Lee, Yin Zhang, Yanglin Zhang, Haonan Wu, Zishan Liu, Ruoxi Zang, Xin Zhu, Jiayin Zheng, Jian Yao, Zefeng Ji, Fei MaarXiv 2026 · DOI

BibTeX
@misc{peng2026be,
  author = {Peng and Lee and Yin Zhang and Yanglin Zhang and Haonan Wu and Zishan Liu and Ruoxi Zang and Xin Zhu and Jiayin Zheng and Jian Yao and Zefeng Ji and Fei Ma},
  title = {{Be Faithful When Response: Returning Fluent and Grounded Answers for Vision-Language Models Reinforcement Learning}},
  year = {2026},
  eprint = {2606.29984},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.29984},
}

Loading related content...

Figure