TIGER-AI-Lab / TIGER-AI-Lab/VLM2Vec

More Retriever Implementations for Latest Multi-Vector Models

Open
#161 2 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
684
Forks
64
Avg merge
8m
Merged PRs (30d)
1

Description

Hi VLM2Vec Team @memray @bryant1410 @wenhuchen ,

Thanks for your great work VLM2Vec-V2 model and MMEB pipeline! I found there are some latest multi-vector models with good performance in the popular vidore-benchmark-v2 (https://huggingface.co/spaces/vidore/vidore-leaderboard) as follows. I believe they also deserve the evaluation on your MMEB benchmark:

I wonder if you can implement these retrievers based on your VLM2Vec-V2 codebase and evaluate them. The community will look forward to this. Thanks!

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

Start by reading the VLM2Vec-V2 codebase and its MMEB benchmark entry points to understand how retrievers are implemented and evaluated. Add support for vidore/colqwen2.5-v0.2, nvidia/llama-nemoretriever-colembed-3b-v1, and nomic-ai/colnomic-embed-multimodal-3b, then run the benchmark and confirm evaluation results for all three models.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface, python
Domain
machine-learning, search
Issue type
Feature
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.