AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

UniTranslator: A Unified Multi-modal framework for End-to-end In-Image Machine Translation

Jiahao Lyu, Pei Fu, Zhenhang Li, Shaojie Zhang, Jiahui Yang, Yu ZHOU, Can Ma, Zhenbo Luo, Jian Luan

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add UGAM and SMD to the baseline (ablation)UniTranslator · Translatotron-V De->EnStructure-BLEU: 9.00 -> 24.86Improved
add UGAM alone (ablation)UniTranslator · Translatotron-V De->EnStructure-BLEU: 9.00 -> 14.28Improved
add UGAM alone (ablation)UniTranslator · Translatotron-V De->EnSSIM: 0.8081 -> 0.8036Hurt
add SMD alone (ablation)UniTranslator · Translatotron-V De->EnStructure-BLEU: 9.00 -> 16.02Improved
add SMD alone (ablation)UniTranslator · Translatotron-V De->EnSSIM: 0.8081 -> 0.8138Improved
add Stage 2 joint fine-tuning to Stage 1 only (ablation)UniTranslator · Translatotron-V De->EnBLEU: 12.53 -> 25.03Improved
add Stage 1 warm-up before Stage 2 (ablation)UniTranslator · Translatotron-V De->EnBLEU: 15.08 -> 25.03Improved
add Stage 1 warm-up before Stage 2 (ablation)UniTranslator · Translatotron-V De->EnSSIM: 0.8148 -> 0.8184Improved
add understanding loss to generation training (ablation)UniTranslator · Translatotron-VStructure-BLEU: 16.57 -> 25.03Improved
add generation loss to understanding training (ablation)UniTranslator · Translatotron-VBLEU: 25.70 -> 34.71Improved
add generation loss to understanding training (ablation)UniTranslator · Translatotron-VCOMET (%): 80.84 -> 84.61Improved
add UGAM to baseline on scene text editing (ablation)UniTranslator · TextEditBenchSSIM: 0.8709 -> 0.8928Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy