Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update and improve Hungarian translations using Claude 3.5 Sonnet #2549

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

PeterDaveHello
Copy link
Contributor

Description

Leverage Claude 3.5 Sonnet from Anthropic, a frontier LLM model, to enhance translation quality. The previous use of Qwen2.5 7b, a smaller and less advanced AI model, may have affected the quality and user experience(#2164). Typically, relying solely on LLMs for translation without verification is not preferred. However, given these concerns, using a frontier LLM model along with references from other language versions should lead to improved results, even for a language I don't even understand.

I also use the powerful frontier OpenAI O1 and Google Gemini-1.5-Pro for proofreading, and manually confirmed line breaks to facilitate easier contributions and improvements by others.

Checklist

  • I have read the Contribution Guidelines
  • I have performed a self-review of my own code
  • I have attached images of the change if it is UI based
  • I have commented my code, particularly in hard-to-understand areas
  • If my code has heavily changed functionality I have updated relevant docs on Stirling-PDFs doc repo
  • My changes generate no new warnings
  • I have read the section Add New Translation Tags (for new translation tags only)

@dosubot dosubot bot added the size:XS This PR changes 0-9 lines, ignoring generated files. label Dec 22, 2024
@PeterDaveHello
Copy link
Contributor Author

@Frooodle, if you need help with the review, maybe https://coderabbit.ai/ could be helpful.

Reference case: transmission-remote-gui/transgui#1479 (comment)

I'm not affiliated with coderabbit, just thought it may help, from my personal experience.

@Ludy87
Copy link
Contributor

Ludy87 commented Dec 22, 2024

📃 File Check: messages_hu_HU.properties

  1. Test Status:Passed
  2. Test Status:Passed

✅ Overall Check Status: Success

@Frooodle

@Ludy87
Copy link
Contributor

Ludy87 commented Dec 22, 2024

/deploypr

Copy link
Contributor

🚀 PR Test Deployment

Your PR has been deployed for testing!

🔗 Test URL: http://185.252.234.121:2549

This deployment will be automatically cleaned up when the PR is closed.

@Frooodle
Copy link
Member

@PeterDaveHello can you merge latest changes into your branch

@Frooodle Frooodle self-requested a review December 22, 2024 18:37
Leverage Claude 3.5 Sonnet from Anthropic, a frontier LLM model, to
enhance translation quality. The previous use of Qwen2.5 7b, a smaller
and less advanced AI model, may have affected the quality and user
experience(Stirling-Tools#2164). Typically, relying solely on LLMs for translation
without verification is not preferred. However, given these concerns,
using a frontier LLM model along with references from other language
versions should lead to improved results, even for a language I don't
even understand.

I also use the powerful frontier OpenAI O1 and Google Gemini-1.5-Pro for
proofreading, and manually confirmed line breaks to facilitate easier
contributions and improvements by others.
@PeterDaveHello
Copy link
Contributor Author

@Frooodle I rebased on the latest main branch, please take a look.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
size:XS This PR changes 0-9 lines, ignoring generated files. Translation
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants