PAIR-code / PAIR-code/lumi

Feature Request: Add Full-Text Paragraph-by-Paragraph Translation

Open
#107 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
207
Forks
31
Avg merge
3h 33m
Merged PRs (30d)
1

Description

Dear Lumi Development Team,

I'm a user of Lumi and find it incredibly useful for reading and understanding arXiv papers. The AI-augmented annotations and smart highlights have significantly improved my research workflow.
I'd like to suggest a new feature: full-text paragraph-by-paragraph translation. This would be particularly valuable for non-native English speakers who need to read papers in their native language while still referencing the original content.

Image Potential Implementation Ideas: Add a translation toggle/selector (e.g., language dropdown) in the UI Display translations alongside original paragraphs (perhaps in a collapsible panel) Preserve existing formatting (bold, italic, equations) in translated text Maintain citation links and references in translated sections This feature would make Lumi more accessible to international researchers and students, expanding its utility across language barriers.

Thank you for considering this request! I'm happy to provide more details or feedback if needed.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named. Start by locating the paper-reading UI and the existing annotation and smart-highlight flows, then determine how paragraph translations, formatting, equations, citations, and language selection should fit together. Done means a user can toggle a target language and view translated paragraphs alongside the original while preserving the requested references and formatting.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
frontend, internationalization
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.