MMTIT-Bench: A Multilingual and Multi-Scenario Benchmark with Cognition-Perception-Reasoning Guided Text-Image Machine Translation

• URL: http://arxiv.org/abs/2603.23896v1
• Date: Wed, 25 Mar 2026 03:37:12 GMT
• Title: MMTIT-Bench: A Multilingual and Multi-Scenario Benchmark with Cognition-Perception-Reasoning Guided Text-Image Machine Translation
• Abstract summary: We present MMTIT-Bench, a human-verified multilingual and multi-scenario benchmark with 1,400 images spanning fourteen non-English and non-Chinese languages.<n>Beyond benchmarking, we study how reasoning-oriented data design improves translation.<n>We propose Cognition-Perception-Reasoning for Translation (CPR-Trans), a data paradigm that integrates scene cognition, text perception, and translation reasoning within a unified reasoning process.
• Score: 29.093072408848467
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: End-to-end text-image machine translation (TIMT), which directly translates textual content in images across languages, is crucial for real-world multilingual scene understanding. Despite advances in vision-language large models (VLLMs), robustness across diverse visual scenes and low-resource languages remains underexplored due to limited evaluation resources. We present MMTIT-Bench, a human-verified multilingual and multi-scenario benchmark with 1,400 images spanning fourteen non-English and non-Chinese languages and diverse settings such as documents, scenes, and web images, enabling rigorous assessment of end-to-end TIMT. Beyond benchmarking, we study how reasoning-oriented data design improves translation. Although recent VLLMs have begun to incorporate long Chain-of-Thought (CoT) reasoning, effective thinking paradigms for TIMT are still immature: existing designs either cascade parsing and translation in a sequential manner or focus on language-only reasoning, overlooking the visual cognition central to VLLMs. We propose Cognition-Perception-Reasoning for Translation (CPR-Trans), a data paradigm that integrates scene cognition, text perception, and translation reasoning within a unified reasoning process. Using a VLLM-driven data generation pipeline, CPR-Trans provides structured, interpretable supervision that aligns perception with reasoning. Experiments on 3B and 7B models show consistent gains in accuracy and interpretability. We will release MMTIT-Bench to promote the multilingual and multi-scenario TIMT research upon acceptance.

Related papers

• Comparative Evaluation of Machine Translation Systems on Images with Text [0.7933039558471405]
  The study compares three main paradigms: modular pipelines that separate text detection, recognition, and translation; multi-modal large language models (MLLMs) capable of processing both image and text jointly; and an end-to-end model, Translatotron-V, which directly generates translated images.
  arXiv  Detail & Related papers  (2026-05-28T07:06:15Z)
• IMTBench: A Multi-Scenario Cross-Modal Collaborative Evaluation Benchmark for In-Image Machine Translation [19.168097777455177]
  End-to-end In-Image Machine Translation aims to convert text embedded within an image into a target language.<n>Existing IIMT benchmarks are largely synthetic and fail to reflect real-world complexity.<n>We present IMTBench, a new benchmark of 2,500 image translation samples covering four practical scenarios and nine languages.
  arXiv  Detail & Related papers  (2026-03-11T07:42:42Z)
• TowerVision: Understanding and Improving Multilinguality in Vision-Language Models [56.775118098058506]
  TowerVision is a family of open multilingual vision-language models for both image-text and video-text tasks.<n>By incorporating visual and cultural context during fine-tuning, our models surpass existing approaches.<n>To support further research, we publicly release all models, data, and training recipes.
  arXiv  Detail & Related papers  (2025-10-22T17:02:48Z)
• A Benchmark for Multi-Lingual Vision-Language Learning in Remote Sensing Image Captioning [27.350370419751385]
  Remote Sensing Image Captioning (RSIC) is a cross-modal field bridging vision and language, aimed at automatically generating natural language descriptions of features and scenes in remote sensing imagery.<n>Two critical challenges persist: the scarcity of non-English descriptive datasets and the lack of multilingual capability evaluation for models.<n>This paper introduces and analyzes BRSIC, a comprehensive bilingual dataset that enriches three established English RSIC datasets with Chinese descriptions, encompassing 13,634 images paired with 68,170 bilingual captions.
  arXiv  Detail & Related papers  (2025-03-06T16:31:34Z)
• Exploring Vision Language Models for Multimodal and Multilingual Stance Detection [9.079302402271491]
  Social media's global reach amplifies the spread of information, highlighting the need for robust Natural Language Processing tasks.<n>Prior research predominantly focuses on text-only inputs, leaving multimodal scenarios relatively underexplored.<n>This paper evaluates state-of-the-art Vision-Language Models (VLMs) on multimodal and multilingual stance detection tasks.
  arXiv  Detail & Related papers  (2025-01-29T13:39:53Z)
• P-MMEval: A Parallel Multilingual Multitask Benchmark for Consistent Evaluation of LLMs [84.24644520272835]
  We introduce P-MMEval, a large-scale benchmark covering effective fundamental and capability-specialized datasets.<n>P-MMEval delivers consistent language coverage across various datasets and provides parallel samples.<n>We conduct extensive experiments on representative multilingual model series to compare performances across models and tasks.
  arXiv  Detail & Related papers  (2024-11-14T01:29:36Z)
• Parrot: Multilingual Visual Instruction Tuning [66.65963606552839]
  Existing methods typically align vision encoders with Multimodal Large Language Models (MLLMs) via supervised fine-tuning (SFT)<n>We propose PARROT, a novel approach that leverages textual guidance for visual token alignment at the language level.<n>We introduce the Massive Multilingual Multimodal Benchmark (MMMB), a new benchmark comprising 6 languages, 15 categories, and 12,000 questions.
  arXiv  Detail & Related papers  (2024-06-04T17:56:28Z)
• Scene Graph as Pivoting: Inference-time Image-free Unsupervised Multimodal Machine Translation with Visual Scene Hallucination [88.74459704391214]
  In this work, we investigate a more realistic unsupervised multimodal machine translation (UMMT) setup. We represent the input images and texts with the visual and language scene graphs (SG), where such fine-grained vision-language features ensure a holistic understanding of the semantics. Several SG-pivoting based learning objectives are introduced for unsupervised translation training. Our method outperforms the best-performing baseline by significant BLEU scores on the task and setup.
  arXiv  Detail & Related papers  (2023-05-20T18:17:20Z)
• IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languages [87.5457337866383]
  We introduce the Image-Grounded Language Understanding Evaluation benchmark. IGLUE brings together visual question answering, cross-modal retrieval, grounded reasoning, and grounded entailment tasks across 20 diverse languages. We find that translate-test transfer is superior to zero-shot transfer and that few-shot learning is hard to harness for many tasks.
  arXiv  Detail & Related papers  (2022-01-27T18:53:22Z)
• UC2: Universal Cross-lingual Cross-modal Vision-and-Language Pre-training [52.852163987208826]
  UC2 is the first machine translation-augmented framework for cross-lingual cross-modal representation learning. We propose two novel pre-training tasks, namely Masked Region-to-Token Modeling (MRTM) and Visual Translation Language Modeling (VTLM) Our proposed framework achieves new state-of-the-art on diverse non-English benchmarks while maintaining comparable performance to monolingual pre-trained models on English tasks.
  arXiv  Detail & Related papers  (2021-04-01T08:30:53Z)
• Simultaneous Machine Translation with Visual Context [42.88121241096681]
  Simultaneous machine translation (SiMT) aims to translate a continuous input text stream into another language with the lowest latency and highest quality possible. We analyse the impact of different multimodal approaches and visual features on state-of-the-art SiMT frameworks.
  arXiv  Detail & Related papers  (2020-09-15T18:19:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.