I want cache-dit-metrics to support a clip score metric for scoring generated images against text prompts.
I want cache-dit-metrics to support a clip score metric for scoring generated images against text prompts.: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). I want cache-dit-metrics to support a clip score metric for scoring generated images against text prompts. The metric…
The task
I want cache-dit-metrics to support a clip_score metric for scoring generated images against text prompts. The metric should appear in cache-dit-metrics --help as a valid positional metric choice, and the prompt flags --prompt-true/-p and --ref-prompt-true/-rp should describe…
Part of XiaomiMiMo/MiMo-V2.6-RL-oss.