Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 

Repository files navigation

MMSep

Implementation instructions and details for MMSep

Install

Follow your desired MLLM's repository for installation.

Usage

Copy and paste functions in mm_separators.py to your desired MLLM's Language model class and call in layered transformer loop. Copy and paste cache.py to your language model codebase, and replace the original Cache class with MMSepCache.

Eval

Collect answers in LLaVA-v1.5 validation jsonl manner and use auto_eval.py with your own evaluation model api.

More details are still under preperation.

About

Implementation instructions and details for KDD 2026 paper MMSep

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages