arXiv:2604.21277v1 Announce Type: new Abstract: We introduce MMTR-Bench, a benchmark designed to evaluate the intrinsic ability of Multimodal Large Language Models (MLLMs) to reconstruct masked text directly from visual context. Unlike conventional question-answering tasks, MMTR-Bench eliminates explicit prompts, requiring models to recover masked text from single- or multi-page inputs across real-world domains such as documents and webpages. This design isolates the reconstruction task from ins
Full text archived locally
✦ AI Summary· Claude Sonnet
Computer Science > Artificial Intelligence
[Submitted on 23 Apr 2026]
Can MLLMs "Read" What is Missing?
Jindi Guo, Xi Fang, Chaozheng Huang
We introduce MMTR-Bench, a benchmark designed to evaluate the intrinsic ability of Multimodal Large Language Models (MLLMs) to reconstruct masked text directly from visual context. Unlike conventional question-answering tasks, MMTR-Bench eliminates explicit prompts, requiring models to recover masked text from single- or multi-page inputs across real-world domains such as documents and webpages. This design isolates the reconstruction task from instruction-following abilities, enabling a direct assessment of a model's layout understanding, visual grounding, and knowledge integration. MMTR-Bench comprises 2,771 test samples spanning multiple languages and varying target lengths. To account for this diversity, we propose a level-aware evaluation protocol. Experiments on representative MLLMs show that the benchmark poses a significant challenge, especially for sentence- and paragraph-level reconstruction. The homepage is available at this https URL.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2604.21277 [cs.AI]
(or arXiv:2604.21277v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2604.21277
Focus to learn more
Submission history
From: Jindi Guo [view email]
[v1] Thu, 23 Apr 2026 04:44:25 UTC (43,411 KB)
Access Paper:
HTML (experimental)
view license
Current browse context:
cs.AI
< prev | next >
new | recent | 2026-04
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Export BibTeX Citation
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Demos
Related Papers
About arXivLabs
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)