modelscope / modelscope/ms-swift
多轮对话DPO训练集构建
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
{"messages": [{"role": "user", "content": "<image>这是什么"}, {"role": "assistant", "content": "这是一只小猫咪。"}], "images": ["cat.png"], "rejected_response": "这是一只小狗。"}
是需要为每一轮对话assistant回答设置rejected_response，即rejected_response为列表，列表元素个数等于对话轮数？
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names no file, test, or entry point. Start by locating the DPO dataset format and preprocessing code for multi-turn conversations, then confirm whether rejected_response accepts one value or a list matching assistant turns; done means the expected format is clearly documented or supported.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100