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

ACL ARR 2026 January Submission6640 Authors

Keywords: Tibetan NLP, low-resource languages, vision-language models, multimodal benchmarks, dataset construction, reproducible evaluation

Abstract:

Vision–language models (VLMs) have progressed rapidly, but Tibetan remains largely underserved due to the lack of infrastructure for reproducible training and evaluation. To help address this gap, we introduce FTibSuite, a resource-centric foundation for Tibetan VLM research that provides an end-to-end training-and-evaluation workflow and includes human-verified multimodal annotations, partially filling a long-standing shortage of Tibetan multimodal resources. FTibSuite comprises FTibData, FTibBench, and a reproducible baseline model, FTibVLM, built on Qwen3-VL-8B-Instruct. FTibVLM adopts a three-stage adaptation pipeline consisting of Tibetan continual pretraining, image–text alignment, and multimodal instruction tuning.
For systematic evaluation, FTibBench adapts five established multimodal benchmarks to Tibetan and offers a reproducible evaluation protocol to support consistent comparisons across models. Specifically, FTibBench includes Tibetan versions of MMBench, MME, POPE, BinaryVQA, and COREVQA. Experiments on FTibBench demonstrate that FTibVLM consistently improves Tibetan multimodal performance. For instance, FTibVLM attains 76.01 accuracy on BinaryVQA, indicating that Tibetan performance can be competitive with high-resource settings on this diagnostic task. We also observe substantial gains on other benchmarks, including an improvement on MMBench (dev) from 42.97 to 67.78 and an increase in POPE-random accuracy from 47.53 to 80.56, underscoring the practical value of the proposed workflow and resources.

Paper Type: Long

Research Area: Multilinguality and Language Diversity

Research Area Keywords: Efficient/Low-Resource Methods for NLP,Efficient/Low-Resource Methods for NLP,Multimodality and Language Grounding to Vision,Robotics and Beyond，Resources and Evaluation

Contribution Types: Reproduction study, Approaches to low-resource settings, Data resources

Languages Studied: Tibetan, Chinese, English

Submission Number: 6640