Elamur Research⌕
☰
home / papers / 2026 / G-Cap: A Game Character Caption Generator

G-Cap: A Game Character Caption Generator

Paper · 2026 · Yang Yang, Feng Hu, Haiming Zhang et al.

Summary

Most captioning research with Large Vision-Language Models (LVLMs) targets everyday photographs. Stylized virtual imagery, such as game and anime characters, gets little attention, even though game studios need such captions for retrieval, skin recommendation, design and content review. The authors define a new task, Game Character Captioning. They build GC-Bench, a human-annotated benchmark, and introduce Graph-F1, a metric that parses captions into graphs. It aligns parent nodes by cosine similarity and uses a reranker to score the attributes and relations attached to them.

Tests on GC-Bench show that even top closed models like Gemini 3 Pro and GPT-5.1 lose ground on this domain, and open-source models trail them further. To narrow that gap, the authors build GC-148K. It combines captions distilled from Gemini 2.5 Pro with synthetic images made by recombining garment components. Fine-tuning Qwen3-VL on this data produces the G-Cap models, which match strong commercial systems at lower cost.

Key points

  • GC-Bench contains 186 manually captioned character images from 20 games, selected for variety in art style and resolution.
  • GC-148K pipeline:
    • About 78K real images were captioned by Gemini 2.5 Pro.
    • Qwen3-235B-A22B-FP8 extracted garment parts across five body regions.
    • A "Fashion Simulator" randomly recombined these parts into new captions.
    • Gemini 2.5 Flash Image (Nano Banana) rendered the new captions as roughly 70K synthetic images.
  • Fine-tuning gives G-Cap-8B and G-Cap-32B, built on Qwen3-VL-8B/32B-Instruct:
    • Graph-F1 rises from 26.09 to 31.45 for the 8B model (about 20.5%) and from 28.78 to 32.31 for the 32B model.
    • G-Cap-32B performs on par with Gemini 3 Pro.
  • Ablation: training on synthetic data alone helps little because of a domain gap, while combining real and synthetic data works best.
  • GPT-5.1 writes the longest captions (over 500 words) but has the lowest information density (0.49).

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 3The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksComputer Vision and Pattern Recognition · 3,486 works

Cite

Cite as Yang Yang, Feng Hu, Haiming Zhang, XU Cheng, Gui Zheng, Liang Yao and Wenqi Ren. 2026. G-Cap: A Game Character Caption Generator. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.248

@inproceedings{yang2026g,
  title = {{G-Cap: A Game Character Caption Generator}},
  author = {Yang Yang, Feng Hu, Haiming Zhang, XU Cheng, Gui Zheng, Liang Yao and Wenqi Ren},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.248},
  url = {https://doi.org/10.18653/v1/2026.acl-long.248}
}

References

38 references, 23 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. 2016. Spice: Semantic propositional image caption evaluation. In European conference on computer vision, pages 382–398. Springer.
  2. J Bai, S Bai, S Yang, S Wang, S Tan, P Wang, J Lin, C Zhou, and J Qwen-VL Zhou. 2023. A versatile vision-language model for understanding, localization, text reading, and beyond. arXiv preprint arXiv:2308.12966, 6.
  3. Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, Wenbin Ge, Zhifang Guo, Qidong Huang, Jie Huang, Fei Huang, Binyuan Hui, Shutong Jiang, Zhaohai Li, Mingsheng Li, and 45 others. 2025a. Qwen3-vl technical report. arXiv prepr…
  4. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, and 1 others. 2025b. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923. — Qwen2.5-VL Technical Report
  5. Satanjeev Banerjee and Alon Lavie. 2005. Meteor: An automatic metric for mt evaluation with improved correlation with human judgments. In Proceedings of the acl workshop on intrinsic and extrinsic evaluation measures for machine translation and/or summarization, pages 65–72.
  6. Yonatan Bitton, Hritik Bansal, Jack Hessel, Rulin Shao, Wanrong Zhu, Anas Awadalla, Josh Gardner, Rohan Taori, and Ludwig Schmidt. 2023. Visit-bench: A benchmark for vision-language instruction following inspired by real-world use. arXiv preprint arXiv:2308.06595. — [arXiv:2308.06595](https://arxiv.org/abs/2308.06595) find
  7. Kanzhi Cheng, Wenpo Song, Jiaxin Fan, Zheng Ma, Qiushi Sun, Fangzhi Xu, Chenyang Yan, Nuo Chen, Jianbing Zhang, and Jiajun Chen. 2025. Caparena: Benchmarking and analyzing detailed image captioning in the llm era. arXiv preprint arXiv:2503.12329. find
  8. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, and 3416 other…
  9. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies,…
  10. Hongyuan Dong, Jiawen Li, Bohong Wu, Jiacong Wang, Yuan Zhang, and Haoyuan Guo. 2024. Benchmarking and improving detail image caption. arXiv preprint arXiv:2405.19092. — [arXiv:2405.19092](https://arxiv.org/abs/2405.19092) find
  11. Yunhao Ge, Xiaohui Zeng, Jacob Samuel Huffman, Tsung-Yi Lin, Ming-Yu Liu, and Yin Cui. 2024. Visual fact checker: Enabling high-fidelity detailed caption generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14033–14042. find
  12. Jack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras, and Yejin Choi. 2021. Clipscore: A reference-free evaluation metric for image captioning. In Proceedings of the 2021 conference on empirical methods in natural language processing, pages 7514– 7528.
  13. Hang Hua, Qing Liu, Lingzhi Zhang, Jing Shi, Soo Ye Kim, Zhifei Zhang, Yilin Wang, Jianming Zhang, Zhe Lin, and Jiebo Luo. 2025. Finecaption: Compositional image captioning focusing on wherever you want at any granularity. In Proceedings of the Computer Vision and Pattern Recognition Conference, pag… find
  14. Andrej Karpathy and Li Fei-Fei. 2015. Deep visual-semantic alignments for generating image descriptions. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3128– 3137.
  15. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph E. Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient memory management for large language model serving with pagedattention. In Proceedings of the ACM SIGOPS 29th Symposium on Operating Systems Principles.
  16. Yebin Lee, Imseong Park, and Myungjoo Kang. 2024. Fleur: An explainable reference-free evaluation metric for image captioning using a large multimodal model. arXiv preprint arXiv:2406.06004. find
  17. Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Peiyuan Zhang, Yanwei Li, Ziwei Liu, and 1 others. 2024. Llava-onevision: Easy visual task transfer. arXiv preprint arXiv:2408.03326.
  18. Xiangtai Li, Tao Zhang, Yanwei Li, Haobo Yuan, Shihao Chen, Yikang Zhou, Jiahao Meng, Yueyi Sun, Shilin Xu, Lu Qi, and 1 others. 2025. Denseworld-1m: Towards detailed dense grounded caption in the real world. arXiv preprint arXiv:2506.24102. — [arXiv:2506.24102](https://arxiv.org/abs/2506.24102) find
  19. Chin-Yew Lin. 2004. Rouge: A package for automatic evaluation of summaries. In Text summarization branches out, pages 74–81.
  20. Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. 2024. Improved baselines with visual instruction tuning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 26296–26306.
  21. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023. Visual instruction tuning. Advances in neural information processing systems, 36:34892– 34916.
  22. Kwai Keye Team. 2025. Kwai keye-vl-1.5 technical report. find
  23. V Team, Wenyi Hong, Wenmeng Yu, Xiaotao Gu, Guo Wang, Guobing Gan, Haomiao Tang, Jiale Cheng, Ji Qi, Junhui Ji, Lihang Pan, Shuaiqi Duan, Weihan Wang, Yan Wang, Yean Cheng, Zehai He, Zhe Su, Zhen Yang, Ziyang Pan, and 69 others. 2025. Glm-4.5v and glm-4.1v-thinking: Towards versatile multimodal reas…
  24. Ramakrishna Vedantam, C Lawrence Zitnick, and Devi Parikh. 2015. Cider: Consensus-based image description evaluation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4566–4575.
  25. Emanuele Vivoli, Niccolò Biondi, Marco Bertini, and Dimosthenis Karatzas. 2024. Comicap: A vlms pipeline for dense captioning of comic panels. In European Conference on Computer Vision, pages 56– 67. Springer. find
  26. Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, and 1 others. 2024. Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution. arXiv preprint arXiv:2409.12191.
  27. George A Miller. 1995. Wordnet: a lexical database for english. Communications of the ACM, 38(11):39–41.
  28. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, and 1 others. 2025. Internvl3.5: Advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265.
  29. Yasumasa Onoe, Sunayana Rane, Zachary Berger, Yonatan Bitton, Jaemin Cho, Roopal Garg, Alexander Ku, Zarana Parekh, Jordi Pont-Tuset, Garrett Tanzer, and 1 others. 2024. Docci: Descriptions of connected and contrasting images. In European Conference on Computer Vision, pages 291–309. Springer. find
  30. Long Xing, Qidong Huang, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Yuhang Cao, Jinsong Li, Shuangrui Ding, Weiming Zhang, Nenghai Yu, and 1 others. 2025. Scalecap: Inference-time scalable image captioning via dual-modality debiasing. arXiv preprint arXiv:2506.19848. — [arXiv:2506.19848](https://arxiv.org/abs/2506.19848) find
  31. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting of the Association for Computational Linguistics, pages 311–318.
  32. Anna Rohrbach, Lisa Anne Hendricks, Kaylee Burns, Trevor Darrell, and Kate Saenko. 2018. Object hallucination in image captioning. arXiv preprint arXiv:1809.02156.
  33. Oriane Siméoni, Huy V. Vo, Maximilian Seitzer, Federico Baldassarre, Maxime Oquab, Cijo Jose, Vasil Khalidov, Marc Szafraniec, Seungeun Yi, Michaël Ramamonjisoa, Francisco Massa, Daniel Haziza, Luca Wehrstedt, Jianyuan Wang, Timothée Darcet, Théo Moutakanni, Leonel Sentana, Claire Roberts, Andrea Ve… — [arXiv:2508.10104](https://arxiv.org/abs/2508.10104) find
  34. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, and 41 others. 2025. Qwen3 technical report. arXiv preprint arXiv:2505.09388.
  35. Yipeng Yu. 2026. Deep research of deep research: From transformer to agent, from ai to ai for science. Preprint, arXiv:2603.28361. — [arXiv:2603.28361](https://arxiv.org/abs/2603.28361) find
  36. Lin Zhang, Xianfang Zeng, Kangcong Li, Gang Yu, and Tao Chen. 2025a. Sc-captioner: Improving image captioning with self-correction by reinforcement learning. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 23145– 23155. find
  37. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q Weinberger, and Yoav Artzi. 2019. Bertscore: Evaluating text generation with bert. arXiv preprint arXiv:1904.09675. find
  38. Yanzhao Zhang, Mingxin Li, Dingkun Long, Xin Zhang, Huan Lin, Baosong Yang, Pengjun Xie, An Yang, Dayiheng Liu, Junyang Lin, and 1 others. 2025b. Qwen3 embedding: Advancing text embedding and reranking through foundation models. arXiv preprint arXiv:2506.05176.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yang-2026-g-cap-game-character-caption/