lm-sys / lm-sys/FastChat

Incorrect Output of Umlauts in deepseek-llm-67b-chat Model

Open
#2,993 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

The deepseek-llm-67b-chat model on is not outputting umlauts correctly, which makes it possible to identify the model. This issue leads to a distorted leaderboard.

Steps to Reproduce:

  1. Use the deepseek-llm-67b-chat model.
  2. Engage in a conversation that includes umlauts (ä, ö, ü) as output.

Expected behavior:
Umlauts should be displayed correctly.

Actual behavior:
Umlauts are displayed incorrectly in the chat output.
Example: "Wir m�chten" instead of "Wir möchten".

Additional information:
This issue affects the integrity of the leaderboard as it allows users to identify the model based on the incorrect output of umlauts.

Contributor guide

No contributing guide indexed for this repository

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 file, test, or entry point is named in the issue. Start by reproducing the umlaut output with deepseek-llm-67b-chat and trace where the chat output is encoded or displayed. Done means ä, ö, and ü render correctly without making the model identifiable through corrupted output.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.