Extend tag-aware translation to hierarchical document formats

Extend the tag-aware translation framework beyond paired HTML/XML tags to hierarchical formats such as LaTeX, particularly when cross-document references must be handled.

Background

The paper’s experiments focus exclusively on paired HTML/XML tags, whose structure differs from hierarchical document formats such as LaTeX. The authors identify support for hierarchical formats involving cross-document references as an unresolved direction for future work, indicating that the current framework’s applicability to such structures has not been established.

References

Extending the framework to handle hierarchical formats such as LaTeX, particularly with cross-document references, remains an open direction for future work.

— Tag-Aware Structured Text Translation: Towards a Systematic Understanding  (2609.29131 - Wu et al., 24 Sep 2026) in Section titled “Limitations,” subsection “Tag type scope”