Elamur Research⌕
☰
home / papers / 2026 / Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework

Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework

Paper · 2026 · Grzegorz Statkiewicz, Alicja Dobrzeniecka, Karolina Seweryn et al.

Summary

Most vision-language models learn mainly from English data, so they handle Polish language and culture poorly. The authors, from NASK, apply the LLaVA-NeXT recipe to Polish and keep human annotation of training data to a minimum. They machine-translate existing multimodal pre-training and instruction datasets with Tower+ 72B and filter the output automatically using COMET-based quality estimation. Synthetic Polish data covers OCR and culture-specific tasks that do not translate well. The resulting models pair a SigLIP2 vision encoder with Polish-native LLMs (PLLuM, Bielik) through a two-layer MLP projector. Human effort went only into evaluation: a carefully revised Polish version of MMBench-dev. The best model beats LLaVA-1.6-Vicuna-13B by 9.5% on Polish MMBench. Human judges also rate its Polish captions as more linguistically correct than those of strong open models. The work suggests that translation plus light filtering can quickly produce usable multimodal models for lower-resource languages. Cultural coverage and OCR-specific evaluation remain open gaps.

Key points

  • Models: the LLaVA-PLLuM and LLaVA-Bielik family (11–12B) uses two PLLuM-12B-nc-instruct variants and Bielik-11B-v2.6-Instruct as backbones, with SigLIP2 So400m/14 (384px) in place of CLIP.
  • Training: the first stage trains only the projector on image–caption pairs. The second stage tunes the projector and vision encoder, while the LLM gets LoRA adapters (r = 128, α = 256).
  • Benchmark: human revision of the translated MMBench-dev found problems in about 4% of samples. The Polish MMBench-dev and the models are released in the NASK-PIB Hugging Face collection.
  • Human evaluation: on an XM3600 subset, the models win on linguistic correctness against PaliGemma2-10B, Pixtral-12B and Qwen2.5-VL-7B at least 64% of the time. The top result is 84% for LLaVA-PLLuM-12b-nc versus Pixtral-12B.
  • Limitations: on content description, the models still trail Qwen2.5-VL-7B. LLM judges missed linguistic differences that humans caught, and key design choices were not tested in ablations.

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 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 works

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 3,320 worksVision & language · 657 works

Cite

Cite as Grzegorz Statkiewicz, Alicja Dobrzeniecka, Karolina Seweryn, Aleksandra Krasnodębska, Karolina Piosek, Katarzyna Bogusz, Sebastian Cygert and Wojciech Kusa. 2026. Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 4: Student Research Workshop).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.eacl-srw.44

@inproceedings{statkiewicz2026annotation,
  title = {{Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework}},
  author = {Grzegorz Statkiewicz, Alicja Dobrzeniecka, Karolina Seweryn, Aleksandra Krasnodębska, Karolina Piosek, Katarzyna Bogusz, Sebastian Cygert and Wojciech Kusa},
  booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 4: Student Research Workshop)},
  year = {2026},
  doi = {10.18653/v1/2026.eacl-srw.44},
  url = {https://doi.org/10.18653/v1/2026.eacl-srw.44}
}

References

45 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Manoj Acharya, Kushal Kafle, and Christopher Kanan. 2019. Tallyqa: answering complex counting questions. In Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence and Thirty-First Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educationa… find
  2. 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.
  3. Pravesh Agrawal, Szymon Antoniak, Emma Bou Hanna, Baptiste Bout, Devendra Chaplot, Jessica Chudnovsky, Diogo Costa, Baudouin De Monicault, Saurabh Garg, Théophile Gervet, Soham Ghosh, Amélie Héliou, Paul Jacob, Albert Q. Jiang, Kartik Khandelwal, Timothée Lacroix, Guillaume Lample, Diego Las Casas,… find
  4. Nahid Alam, Karthik Reddy Kanjula, Surya Guthikonda, Timothy Chung, Bala Krishna S Vegesna, Abhipsha Das, Anthony Susevski, Ryan Sze-Yin Chan, S M Iftekhar Uddin, Shayekh Bin Islam, Roshan Santhosh, Snegha A, Drishti Sharma, Chen Liu, Isha Chaturvedi, Genta Indra Winata, Ashvanth. S, Snehanshu Mukhe… — [arXiv:2412.07112](https://arxiv.org/abs/2412.07112) find
  5. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katie Millicah, Malcolm Reynolds, Roman Ring, Eliza Rutherford, Serkan Cabi, Tengda Han, Zhitao Gong, Sina Samangooei, Marianne Monteiro, Jacob Menick, Sebastian Borgeaud, and 8 others… — Flamingo: a Visual Language Model for Few-Shot Learning
  6. Lucas Beyer, Andreas Steiner, André Susano Pinto, Alexander Kolesnikov, Xiao Wang, Daniel Salz, Maxim Neumann, Ibrahim Alabdulmohsin, Michael Tschannen, Emanuele Bugliarello, Thomas Unterthiner, Daniel Keysers, Skanda Koppula, Fangyu Liu, Adam Grycner, Alexey Gritsenko, Neil Houlsby, Manoj Kumar, Ke… find
  7. Guiming Hardy Chen, Shunian Chen, Ruifei Zhang, Junying Chen, Xiangbo Wu, Zhiyi Zhang, Zhihong Chen, Jianquan Li, Xiang Wan, and Benyou Wang. 2024. Allava: Harnessing gpt4v-synthesized data for lite vision-language models. Preprint, arXiv:2402.11684.
  8. Alexis Conneau and Guillaume Lample. 2019. Cross-lingual language model pretraining. In Proceedings of the 33rd International Conference on Neural Information Processing Systems, Red Hook, NY, USA. Curran Associates Inc. find
  9. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. LoRA: Low-rank adaptation of large language models. In International Conference on Learning Representations.
  10. Geewook Kim, Teakgyu Hong, Moonbin Yim, JeongYeon Nam, Jinyoung Park, Jinyeong Yim, Wonseok Hwang, Sangdoo Yun, Dongyoon Han, and Seunghyun Park. 2022. Ocr-free document understanding transformer. In European Conference on Computer Vision (ECCV).
  11. Jan Kocoń, Maciej Piasecki, Arkadiusz Janz, Teddy Ferdinan, Łukasz Radliński, Bartłomiej Koptyra, Marcin Oleksy, Stanisław Woźniak, Paweł Walkowiak, Konrad Wojtasik, Julia Moska, Tomasz Naskr˛et, Bartosz Walkowiak, Mateusz Gniewkowski, Kamil Szyc, Dawid Motyka, Dawid Banach, Jonatan Dalasiński,… — [arXiv:2511.03823](https://arxiv.org/abs/2511.03823) find
  12. 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.
  13. Junnan Li, Ramprasaath Selvaraju, Akhilesh Gotmare, Shafiq Joty, Caiming Xiong, and Steven Chu Hong Hoi. 2021. Align before fuse: Vision and language representation learning with momentum distillation. Advances in neural information processing systems, 34:9694–9705. find
  14. Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C. Lawrence Zitnick. 2014. Microsoft coco: Common objects in context. In Computer Vision – ECCV 2014, pages 740–755, Cham. Springer International Publishing.
  15. Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. 2024a. Improved baselines with visual instruction tuning. In 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 26286–26296.
  16. Haotian Liu, Chunyuan Li, Yuheng Li, Bo Li, Yuanhan Zhang, Sheng Shen, and Yong Jae Lee. 2024b. Llava-next: Improved reasoning, ocr, and world knowledge. find
  17. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023. Visual instruction tuning. In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Hook, NY, USA. Curran Associates Inc.
  18. Yuan Liu, Haodong Duan, Yuanhan Zhang, Bo Li, Songyang Zhang, Wangbo Zhao, Yike Yuan, Jiaqi Wang, Conghui He, Ziwei Liu, Kai Chen, and Dahua Lin. 2024c. Mmbench: Is your multi-modal model an all-around player? In European Conference on Computer Vision (ECCV). Oral Presentation; arXiv:2307.06281.
  19. Hugo Laurençon, Léo Tronchon, Matthieu Cord, and Victor Sanh. 2024. What matters when building vision-language models? In Proceedings of the 38th International Conference on Neural Information Processing Systems, NIPS ’24, Red Hook, NY, USA. Curran Associates Inc.
  20. Elio Musacchio, Lucia Siciliani, Pierpaolo Basile, and Giovanni Semeraro. 2024. Llava-ndino: Empowering llms with multimodality for the italian language. In Proceedings of the Eighth Workshop on Natural Language for Artificial Intelligence (NL4AI 2024), co-located with the 23rd International Confere… find
  21. Junchen Li, Qing Yang, Bojian Jiang, Shaolin Zhu, and Qingxuan Sun. 2025. Lrm-llava: Overcoming the modality gap of multilingual large language-vision model for low-resource languages. Proceedings of the AAAI Conference on Artificial Intelligence, 39(23):24449–24457. find
  22. Krzysztof Ociepa, Krzysztof Wróbel, Adrian Gwoździej, Remigiusz Kinas, and 1 others. 2025. Bielik 11b v2 technical report. arXiv preprint arXiv:2505.02410. — [arXiv:2505.02410](https://arxiv.org/abs/2505.02410) find
  23. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023. 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.
  24. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, and 1 others. 2022. Training language models to follow instructions with human feedback. Advances in neural information processing systems, 35:27730–27744.
  25. Piotr P˛ezik, Filip Żarnecki, Konrad Kaczyński, Anna Cichosz, Zuzanna Deckert, Monika Garnys, Izabela Grabarczyk, Wojciech Janowski, Sylwia Karasińska, Aleksandra Kujawiak, Piotr Misztela, Maria Szy-mańska, Karolina Walkusz, Igor Siek, Maciej Chrab aszcz, Anna Kołos, Agnieszka Karlińska, Karol… — [arXiv:2511.17161](https://arxiv.org/abs/2511.17161) find
  26. Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. 2021. Learning transferable visual models from natural language supervision. In Proceedings of the 38th Internatio…
  27. Ricardo Rei, Nuno M. Guerreiro, José Pombal, João Alves, Pedro Teixeirinha, Amin Farajian, and André F. T. Martins. 2025. Tower+: Bridging generality and translation specialization in multilingual llms. Preprint, arXiv:2506.17080. find
  28. Ricardo Rei, Craig Stewart, Ana C Farinha, and Alon Lavie. 2020. Unbabel’s participation in the WMT20 metrics shared task. In Proceedings of the Fifth Conference on Machine Translation, pages 911–920, Online. Association for Computational Linguistics.
  29. Dustin Schwenk, Apoorv Khandelwal, Christopher Clark, Kenneth Marino, and Roozbeh Mottaghi. 2022. A-okvqa: A benchmark for visual question answering using world knowledge. In Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part VIII, page 14…
  30. Karolina Seweryn, Anna Kołos, Agnieszka Karlińska, Katarzyna Lorenc, Katarzyna Dziewulska, Maciej Chrabaszcz, Aleksandra Krasnod˛ebska, Paula Betscher, Zofia Cieślińska, Katarzyna Kowol, and 1 others. 2025. Pllum-align: Polish preference dataset for large language model alignment. In Proceedings… find
  31. DongJae Shin, HyeonSeok Lim, Inho Won, ChangSu Choi, Minjun Kim, SeungWoo Song, HanGyeol Yoo, SangMin Kim, and KyungTae Lim. 2024. X-LLaVA: Optimizing bilingual large vision-language alignment. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 2463–2473, Mexico City, Me… find
  32. Krishna Srinivasan, Karthik Raman, Jiecao Chen, Michael Bendersky, and Marc Najork. 2021a. Wit: Wikipedia-based image text dataset for multimodal multilingual machine learning. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’… find
  33. Krishna Srinivasan, Karthik Raman, Jiecao Chen, Michael Bendersky, and Marc Najork. 2021b. Wit: Wikipedia-based image text dataset for multimodal multilingual machine learning. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’… find
  34. Andreas Steiner, André Susano Pinto, Michael Tschannen, Daniel Keysers, Xiao Wang, Yonatan Bitton, Alexey Gritsenko, Matthias Minderer, Anthony Sherbondy, Shangbang Long, Siyang Qin, Reeve Ingle, Emanuele Bugliarello, Sahar Kazemzadeh, Thomas Mesnard, Ibrahim Alabdulmohsin, Lucas Beyer, and Xiaohua… — [arXiv:2412.03555](https://arxiv.org/abs/2412.03555) find
  35. Ashish V. Thapliyal, Jordi Pont Tuset, Xi Chen, and Radu Soricut. 2022. Crossmodal-3600: A massively multilingual multimodal evaluation dataset. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 715–729, Abu Dhabi, United Arab Emirates. Association for…
  36. Shengbang Tong, Ellis Brown, Penghao Wu, Sanghyun Woo, Manoj Middepogu, Sai Charitha Akula, Jihan Yang, Shusheng Yang, Adithya Iyer, Xichen Pan, Austin Wang, Rob Fergus, Yann LeCun, and Saining Xie. 2024. Cambrian-1: A fully open, vision-centric exploration of multimodal llms. In Advances in Neural…
  37. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, and 1 others. 2023. LLaMA: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971.
  38. Michael Tschannen, Alexey Gritsenko, Xiao Wang, Muhammad Ferjad Naeem, Ibrahim Alabdulmohsin, Nikhil Parthasarathy, Talfan Evans, Lucas Beyer, Ye Xia, Basil Mustafa, Olivier Hénaff, Jeremiah Harmsen, Andreas Steiner, and Xiaohua Zhai. 2025. Siglip 2: Multilingual vision-language encoders with improv… find
  39. Junke Wang, Lingchen Meng, Zejia Weng, Bo He, Zuxuan Wu, and Yu-Gang Jiang. 2023. To see is to believe: Prompting gpt-4v for better visual instruction tuning. arXiv preprint arXiv:2311.07574. — [arXiv:2311.07574](https://arxiv.org/abs/2311.07574) find
  40. Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Yang Fan, Kai Dang, Mengfei Du, Xuancheng Ren, Rui Men, Dayiheng Liu, Chang Zhou, Jingren Zhou, and Junyang Lin. 2024. Qwen2-vl: Enhancing vision-language model’s perception of the w…
  41. Luis Wiedmann, Orr Zohar, Amir Mahla, Xiaohan Wang, Rui Li, Thibaud Frere, Leandro von Werra, Aritra Roy Gosthipaty, and Andrés Marafioti. 2025. Finevision: Open data is all you need. Preprint, arXiv:2510.17269. — [arXiv:2510.17269](https://arxiv.org/abs/2510.17269) find
  42. Haoning Wu, Zicheng Zhang, Erli Zhang, Chaofeng Chen, Liang Liao, Annan Wang, Kaixin Xu, Chunyi Li, Jingwen Hou, Guangtao Zhai, Geng Xue, Wenxiu Sun, Qiong Yan, and Weisi Lin. 2024. Q-instruct: Improving low-level visual abilities for multi-modality foundation models. In 2024 IEEE/CVF Conference on… find
  43. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, Kai Dang, and 23 others. 2024. Qwen2.5 technical report. arXiv preprint arXi… find
  44. Jingyi Zhang, Jiaxing Huang, Sheng Jin, and Shijian Lu. 2024. Vision-language models for vision tasks: A survey. IEEE transactions on pattern analysis and machine intelligence, 46(8):5625–5644. find
  45. Training data Dataset \# Samples Stage 1 Stage 2 find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/statkiewicz-2026-annotation-efficient-vision-language-model/