Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix: Fixed eval split for MultilingualSentiment in C-MTEB #1804

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: Fixed eval split for MultilingualSentiment in C-MTEB #1804

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
70 changes: 36 additions & 34 deletions mteb/benchmarks/benchmarks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1150,40 +1150,42 @@ def load_results(

C_MTEB = Benchmark(
name="MTEB(Chinese)",
tasks=get_tasks(
tasks=[
"T2Retrieval",
"MMarcoRetrieval",
"DuRetrieval",
"CovidRetrieval",
"CmedqaRetrieval",
"EcomRetrieval",
"MedicalRetrieval",
"VideoRetrieval",
"T2Reranking",
"MMarcoReranking",
"CMedQAv1-reranking",
"CMedQAv2-reranking",
"Ocnli",
"Cmnli",
"CLSClusteringS2S",
"CLSClusteringP2P",
"ThuNewsClusteringS2S",
"ThuNewsClusteringP2P",
"ATEC",
"BQ",
"LCQMC",
"PAWSX",
"STSB",
"AFQMC",
"QBQTC",
"TNews",
"IFlyTek",
"Waimai",
"OnlineShopping",
"MultilingualSentiment",
"JDReview",
],
tasks=MTEBTasks(
get_tasks(
tasks=[
"T2Retrieval",
"MMarcoRetrieval",
"DuRetrieval",
"CovidRetrieval",
"CmedqaRetrieval",
"EcomRetrieval",
"MedicalRetrieval",
"VideoRetrieval",
"T2Reranking",
"MMarcoReranking",
"CMedQAv1-reranking",
"CMedQAv2-reranking",
"Ocnli",
"Cmnli",
"CLSClusteringS2S",
"CLSClusteringP2P",
"ThuNewsClusteringS2S",
"ThuNewsClusteringP2P",
"ATEC",
"BQ",
"LCQMC",
"PAWSX",
"STSB",
Comment on lines +1174 to +1178
Copy link
Collaborator

@Samoed Samoed Jan 14, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Atec, BQ, STSB have multiple splits too

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

and they should only use the test split then?

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It seems that in previous leaderboard validation split was used https://github.com/embeddings-benchmark/results/blob/4f6a9fc42a2e67922d36ad3d41baf3aa9c969e6e/results.py#L236-L246
But I can't find information in their repo/paper

Copy link
Contributor

@KennethEnevoldsen KennethEnevoldsen Jan 15, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let us just do whatever they do on the leaderboard (I believe they also have a test set which is hidden, but we don't have access to that)

"AFQMC",
"QBQTC",
"TNews",
"IFlyTek",
"Waimai",
"OnlineShopping",
"JDReview",
],
)
+ get_tasks(tasks=["MultilingualSentiment"], eval_splits=["test"])
),
description="The Chinese Massive Text Embedding Benchmark (C-MTEB) is a comprehensive benchmark for Chinese text embeddings covering 6 tasks and 35 datasets.",
reference="https://github.com/FlagOpen/FlagEmbedding/tree/master/research/C_MTEB",
Expand Down
Loading