---
task_categories:
- sentence-similarity
language:
- af
- am
- te
- mr
- pa
- hi
- rw
- ha
- id
- es
- arb
- arq
- ary
configs:
  - config_name: afr_Latn
    data_files:
      - split: test
        path: data/afr_Latn/test*
      - split: dev
        path: data/afr_Latn/dev*
  - config_name: amh_Ethi
    data_files:
      - split: test
        path: data/amh_Ethi/test*
      - split: dev
        path: data/amh_Ethi/dev*
  - config_name: arb_Arab
    data_files:
      - split: test
        path: data/arb_Arab/test*
      - split: dev
        path: data/arb_Arab/dev*
  - config_name: arq_Arab
    data_files:
      - split: test
        path: data/arq_Arab/test*
      - split: dev
        path: data/arq_Arab/dev*
  - config_name: ary_Arab
    data_files:
      - split: test
        path: data/ary_Arab/test*
      - split: dev
        path: data/ary_Arab/dev*
  - config_name: eng_Latn
    data_files:
      - split: test
        path: data/eng_Latn/test*
      - split: dev
        path: data/eng_Latn/dev*
  - config_name: hau_Latn
    data_files:
      - split: test
        path: data/hau_Latn/test*
      - split: dev
        path: data/hau_Latn/dev*
  - config_name: hin_Deva
    data_files:
      - split: test
        path: data/hin_Deva/test*
      - split: dev
        path: data/hin_Deva/dev*
  - config_name: hin_Latn
    data_files:
      - split: test
        path: data/hin_Latn/test*
      - split: dev
        path: data/hin_Latn/dev*
  - config_name: ind_Latn
    data_files:
      - split: test
        path: data/ind_Latn/test*
      - split: dev
        path: data/ind_Latn/dev*
  - config_name: kin_Latn
    data_files:
      - split: test
        path: data/kin_Latn/test*
      - split: dev
        path: data/kin_Latn/dev*
  - config_name: mar_Deva
    data_files:
      - split: test
        path: data/mar_Deva/test*
      - split: dev
        path: data/mar_Deva/dev*
  - config_name: pan_Guru
    data_files:
      - split: test
        path: data/pan_Guru/test*
      - split: dev
        path: data/pan_Guru/dev*
  - config_name: spa_Latn
    data_files:
      - split: test
        path: data/spa_Latn/test*
      - split: dev
        path: data/spa_Latn/dev*
  - config_name: tel_Telu
    data_files:
      - split: test
        path: data/tel_Telu/test*
      - split: dev
        path: data/tel_Telu/dev*
---

**Source dataset**:   
- Link: [SemRel/SemRel2024](https://huggingface.co/datasets/SemRel/SemRel2024)  
- Revision: `ef5c383d1b87eb8feccde3dfb7f95e42b1b050dd`

**SemRel2024**:  
SemRel2024 is a collection of Semantic Textual Relatedness (STR) datasets for 14 languages, including African and Asian languages. The datasets are composed of sentence pairs, each assigned a relatedness score between 0 (completely) unrelated and 1 (maximally related) with a large range of expected relatedness values. SemRel2024 dataset was used as part of the SemEval2024 shared task 1. The task aims to evaluate the ability of systems to measure the semantic relatedness between two sentences.

**SemRel2024 Plus**   
SemRel2024 Plus additionally has hin_Latn data generated using indictrans library.