Skip to content

Questions on Training Data Effectiveness for Additional Models #38

Description

@Cuijr27

Thanks for your excellent work!
In your paper, Qwen3-8B was used as the reference model. However, its test pass rate is quite low on the Terminal-Bench-2 benchmark. I was wondering if you have also evaluated the Qwen3.5 series as baselines to assess the performance improvement brought by SFT on the Terminal-Bench-2 benchmark.
Looking forward to your response!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions