open-compass / open-compass/VLMEvalKit

There are some bugs within RefCOCO dataset

Open
#1,330 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
4.4k
Forks
768
Avg merge
1d 10h
Merged PRs (30d)
17

Description

'RefCOCO': "https://huggingface.co/datasets/mjuicem/RefCOCO-VLMEvalKit/resolve/main/RefCOCO.tsv"

e.g.
119 /9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGB8aAP......./2Q== Please provide the bounding box coordinate of the region this sentence describes: rightest guyin black [0.088250, 0.033709, 0.469672, 0.474155] RefCOCO+ test-A
--- I suppose the prompt should be "guy in", not "guyin".
Hi @mjuicem Could you help check this out, and see if there are other issues about the dataset. 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 with the RefCOCO TSV at the linked Hugging Face URL and inspect the shown RefCOCO+ test-A row. Verify whether “guyin” should be “guy in” and check the dataset for similar prompt issues. Done means the confirmed dataset errors are corrected or clearly reported, including any additional issues found.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface, python
Domain
computer-vision, data
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.