Elamur Research⌕
☰
home / papers / 2026 / FTibSuite: A Comprehensive Resource Suite for Tibetan Vision–Language Modeling

FTibSuite: A Comprehensive Resource Suite for Tibetan Vision–Language Modeling

Paper · 2026 · Guixian Xu, Yide Liang, Zeli Su et al.

Summary

Tibetan has been largely left out of recent progress in vision–language models. It lacks image–text data, multimodal instruction sets and standardized benchmarks, so Tibetan VLM results are hard to reproduce or compare. The authors address this with FTibSuite, which has three parts:

  • FTibData: training material for each adaptation step, namely Tibetan text for continual pretraining, Tibetan image–text pairs for alignment, and multimodal instructions.
  • FTibBench: an evaluation suite that ports five established English benchmarks into Tibetan under a common protocol.
  • FTibVLM: a baseline built by adapting Qwen3-VL-8B-Instruct in three stages (continual pretraining, image–text alignment, instruction tuning).

To keep translation errors out of the benchmarks, outputs were checked by an LLM, flagged items were corrected by Tibetan experts, and a sample of accepted items was audited by humans. FTibVLM clearly outperforms the unadapted backbone on Tibetan tasks. Because the training mix includes Chinese data, it keeps its Chinese multimodal skills almost unchanged. The suite gives low-resource multimodal research a reusable starting point.

Key points

  • FTibBench translates five benchmarks into Tibetan: MMBench, MME, POPE, BinaryVQA and COREVQA. These cover hallucination, binary decisions, crowded-scene reasoning, capability profiling and multiple-choice reasoning.
  • Quality control used DeepSeek-V3 for automatic verification and scoring, mandatory expert correction of low-quality items, and human review of 10% of auto-accepted samples.
  • Tibetan results for FTibVLM against the backbone:
    • MMBench rose from 42.97 to 67.78.
    • POPE-random accuracy rose from 47.53 to 80.56.
    • BinaryVQA reached 76.01.
  • Comparing stage-wise checkpoints (Base, CP+MA, FTibVLM) shows steady gains as each stage is added. The largest jump comes from continual pretraining plus alignment.
  • Chinese retention on MMBench-CN (dev) is 88.15 versus 88.50 for Base. The model and data are released on Hugging Face (onedday/FTib-VLM, onedday/FTib-data).

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

 Institutions

🎓Shanghai Jiao Tong UniversityShanghai · China212 works · 19,544 citations🏛️Chinese Academy of SciencesBeijing · China246 works · 24,952 citations🔬Institute of AutomationBeijing · China87 works · 14,883 citations🎓Minzu University of ChinaBeijing · China7 works · 2 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksComputer Vision and Pattern Recognition · 3,486 works

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 3,894 worksVision & language · 811 works

Cite

Cite as Guixian Xu, Yide Liang, Zeli Su, Xuexian Song, Ziyin Zhang, Yushuang Dong, Ting Zhang and Xu Han. 2026. FTibSuite: A Comprehensive Resource Suite for Tibetan Vision–Language Modeling. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.903

@inproceedings{xu2026ftibsuite,
  title = {{FTibSuite: A Comprehensive Resource Suite for Tibetan Vision–Language Modeling}},
  author = {Guixian Xu, Yide Liang, Zeli Su, Xuexian Song, Ziyin Zhang, Yushuang Dong, Ting Zhang and Xu Han},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.903},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.903}
}

References

38 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  2. Nahid Alam, Karthik Reddy Kanjula, Surya Guthikonda, Timothy Chung, Bala Krishna S Vegesna, Abhipsha Das, Anthony Susevski, Ryan Sze-Yin Chan, SM Uddin, Shayekh Bin Islam, and 1 others. 2025. Behind maya: Building a multilingual vision language model. arXiv preprint arXiv:2505.08910. — [arXiv:2505.08910](https://arxiv.org/abs/2505.08910) find
  3. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, and 1 others. 2022. Flamingo: a visual language model for few-shot learning. Advances in neural information processing systems, 35:23716– 23736.
  4. Bo An. 2023. Prompt-based for low-resource tibetan text classification. ACM Transactions on Asian and Low-Resource Language Information Processing, 22(8):1–13. find
  5. Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, and 1 others. 2023. Qwen technical report. arXiv preprint arXiv:2309.16609.
  6. Ali Borji. 2023. Binaryvqa: A versatile test set to evaluate the out-of-distribution generalization of vqa models. arXiv preprint arXiv:2301.12032. — [arXiv:2301.12032](https://arxiv.org/abs/2301.12032) find
  7. Lifeng Chen, Ryan Lai, and Tianming Liu. 2025. Adapting large language models to low-resource tibetan: A two-stage continual and supervised fine-tuning study. arXiv preprint arXiv:2512.03976. find
  8. Xi Chen, Xiao Wang, Soravit Changpinyo, Anthony J Piergiovanni, Piotr Padlewski, Daniel Salz, Sebastian Goodman, Adam Grycner, Basil Mustafa, Lucas Beyer, and 1 others. 2022. Pali: A jointly-scaled multilingual language-image model. arXiv preprint arXiv:2209.06794.
  9. Ishant Chintapatla, Kazuma Choji, Naaisha Agarwal, Andrew Lin, Hannah You, Charles Duong, Kevin Zhu, Sean O’Brien, and Vasu Sharma. 2025. Corevqa: A crowd observation and reasoning entailment visual question answering benchmark. arXiv preprint arXiv:2507.13405. — [arXiv:2507.13405](https://arxiv.org/abs/2507.13405) find
  10. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Proceedings of the 58th annual meeting of the association fo…
  11. Marta R Costa-Jussà, James Cross, Onur Çelebi, Maha Elbayad, Kenneth Heafield, Kevin Heffernan, Elahe Kalbassi, Janice Lam, Daniel Licht, Jean Maillard, and 1 others. 2022. No language left behind: Scaling human-centered machine translation. arXiv preprint arXiv:2207.04672. find
  12. Wenliang Dai, Junnan Li, Dongxu Li, Anthony Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale N Fung, and Steven Hoi. 2023. Instructblip: Towards general-purpose vision-language models with instruction tuning. Advances in neural information processing systems, 36:49250–49267.
  13. DeepSeek-AI, Aixin Liu, Aoxue Mei, Bangcai Lin, Bing Xue, Bingxuan Wang, Bingzheng Xu, Bochao Wu, Bowei Zhang, Chaofan Lin, Chen Dong, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenhao Xu, Chong Ruan, Damai Dai, Daya Guo, Dejian Yang, and 245 others. 2025. Deepseek-v3.2: Pushing the frontier of open…
  14. Chaoyou Fu, Peixian Chen, Yunhang Shen, Yulei Qin, Mengdan Zhang, Xu Lin, Jinrui Yang, Xiawu Zheng, Ke Li, Xing Sun, and 1 others. 2025. Mme: A comprehensive evaluation benchmark for multimodal large language models. In The Thirty-ninth Annual Conference on Neural Information Processing Systems Data…
  15. Suchin Gururangan, Ana Marasović, Swabha Swayamdipta, Kyle Lo, Iz Beltagy, Doug Downey, and Noah A Smith. 2020. Don’t stop pretraining: Adapt language models to domains and tasks. arXiv preprint arXiv:2004.10964. find
  16. Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, and 1 others. 2022. Training compute-optimal large language models. arXiv preprint arXiv:2203.15556.
  17. Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin De Laroussilhe, Andrea Gesmundo, Mona Attariyan, and Sylvain Gelly. 2019. Parameter-efficient transfer learning for nlp. In International conference on machine learning, pages 2790–2799. PMLR. find
  18. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, and 1 others. 2022. Lora: Low-rank adaptation of large language models. ICLR, 1(2):3.
  19. Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. 2020. Scaling laws for neural language models. arXiv preprint arXiv:2001.08361.
  20. Fenfang Li, Zhengzhang Zhao, Li Wang, and Han Deng. 2024. Tibetan sentence boundaries automatic disambiguation based on bidirectional encoder representations from transformers on byte pair encoding word cutting method. Applied Sciences, 14(7):2989. find
  21. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023a. Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. In International conference on machine learning, pages 19730–19742. PMLR.
  22. Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. 2022. Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation. In International conference on machine learning, pages 12888–12900. PMLR.
  23. Yifan Li, Yifan Du, Kun Zhou, Jinpeng Wang, Wayne Xin Zhao, and Ji-Rong Wen. 2023b. Evaluating object hallucination in large vision-language models. arXiv preprint arXiv:2305.10355.
  24. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023. Visual instruction tuning. Advances in neural information processing systems, 36:34892– 34916.
  25. Atsuyuki Miyai, Jingkang Yang, Jingyang Zhang, Yifei Ming, Qing Yu, Go Irie, Yixuan Li, Hai Helen Li, Ziwei Liu, and Kiyoharu Aizawa. 2025. Unsolvable problem detection: Robust understanding evaluation for large multimodal models. In Proceedings of the 63rd Annual Meeting of the Association for Comp… find
  26. Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, and 1 others. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learning, pages 8748–8…
  27. Cheng Huang, Nyima Tashi, Fan Gao, Yutong Liu, Jiahao Li, Hao Tian, Siyang Jiang, Thupten Tsering, Ban Ma-bao, Renzeg Duojie, and 1 others. 2025. Tibetan language and ai: A comprehensive survey of resources, methods and challenges. arXiv preprint arXiv:2510.19144. — [arXiv:2510.19144](https://arxiv.org/abs/2510.19144) find
  28. Zeli Su, Ziyin Zhang, Guixian Xu, Jianing Liu, Xu Han, Ting Zhang, and Yushuang Dong. 2025. Multilingual encoder knows more than you realize: Shared weights pretraining for extremely low-resource languages. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (V… find
  29. Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc Le, Yun-Hsuan Sung, Zhen Li, and Tom Duerig. 2021. Scaling up visual and vision-language representation learning with noisy text supervision. In International conference on machine learning, pages 4904–4916. PMLR.
  30. Jiahong Wu, He Zheng, Bo Zhao, Yixin Li, Baoming Yan, Rui Liang, Wenjia Wang, Shipei Zhou, Guosen Lin, Yanwei Fu, Yizhou Wang, and Yonggang Wang. 2017. AI challenger : A large-scale dataset for going deeper in image understanding. CoRR, abs/1711.06475. — [arXiv:1711.06475](https://arxiv.org/abs/1711.06475) find
  31. Bushi Xiao, Qian Shen, and Daisy Zhe Wang. 2025. From text to multi-modal: Advancing low-resource-language translation through synthetic data generation and cross-modal alignments. In Proceedings of the Eighth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2025),… find
  32. Zhiyang Xu, Chao Feng, Rulin Shao, Trevor Ashby, Ying Shen, Di Jin, Yu Cheng, Qifan Wang, and Lifu Huang. 2024. Vision-flan: Scaling human-labeled tasks in visual instruction tuning. arXiv preprint arXiv:2402.11690. find
  33. 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. Preprint, arXiv:2505.09388.
  34. Ziqing Yang, Zihang Xu, Yiming Cui, Baoxin Wang, Min Lin, Dayong Wu, and Zhigang Chen. 2022. Cino: A chinese minority pre-trained language model. arXiv preprint arXiv:2202.13558. find
  35. Chen Zhang, Mingxu Tao, Quzhe Huang, Jiuheng Lin, Zhibin Chen, and Yansong Feng. 2023. Mcˆ 2: A multilingual corpus of minority languages in china. CoRR. find
  36. Chen Zhang, Mingxu Tao, Quzhe Huang, Jiuheng Lin, Zhibin Chen, and Yansong Feng. 2024. Mc2: Towards transparent and culturally-aware nlp for minority languages in china. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8832–885… find
  37. Duzhen Zhang, Yong Ren, Zhong-Zhi Li, Yahan Yu, Jiahua Dong, Chenxing Li, Zhilong Ji, and Jinfeng Bai. 2025. Enhancing multimodal continual instruction tuning with branchlora. arXiv preprint arXiv:2506.02041. find
  38. Wenzhen Zheng, Wenbo Pan, Xu Xu, Libo Qin, Li Yue, and Ming Zhou. 2024. Breaking language barriers: Cross-lingual continual pre-training at scale. arXiv preprint arXiv:2407.02118. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/xu-2026-ftibsuite-comprehensive-resource-suite-tibetan/