dataset_info:
features:
- name: text
dtype: string
- name: label
list: string
splits:
- name: train
num_bytes: 4898199
num_examples: 6400
- name: test
num_bytes: 1243908
num_examples: 1600
download_size: 3430719
dataset_size: 6142107
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
RuSciBenchZhoMultilabelClassification
The RuSciBenchZhoMultilabelClassification task addresses the assignment of multiple semantic categories to a single scientific article, reflecting the interdisciplinary nature of modern research. The dataset used for this benchmark was collected from sciencechina.cn, where articles are indexed with titles, abstracts, and typically multiple thematic labels.
Originally, the metadata comprised 90 unique labels. To organize these into coherent groups, we utilized DeepSeek v3.1 for clustering, followed by a manual validation step. Labels that were semantically distant from the main groups were aggregated into an "Other" category.
For the final dataset, we excluded the "Other" category and specifically selected articles associated with multiple high-level classes. The data was subsequently split into training and test sets for evaluation.
How to evaluate on this task
First, install MTEB version with this task:
pip install git+https://github.com/mlsa-iai-msu-lab/ru_sci_bench_mteb.git@ruscibench
Then run code evaluate a model on this task:
import mteb
from sentence_transformers import SentenceTransformer
model_name = "sentence-transformers/all-MiniLM-L6-v2"
model = mteb.get_model(model_name)
tasks = mteb.get_tasks(tasks=["RuSciBenchZhoMultilabelClassification"])
results = mteb.evaluate(model, tasks=tasks)