README.md

---
configs:
  - config_name: default
    data_files:
      - split: test
        path: qrels/test.jsonl
  - config_name: corpus
    data_files:
      - split: corpus
        path: corpus.jsonl
  - config_name: queries
    data_files:
      - split: queries
        path: queries.jsonl
---
## Dataset Summary

**CQADupstack-mathematica-Fa** is a Persian (Farsi) dataset created for the **Retrieval** task, focused on **duplicate question detection** in community question-answering (CQA) forums. It is a translated version of the **"mathematica" (Mathematica Stack Exchange)** subforum from the English **CQADupstack** dataset and is part of the **FaMTEB** benchmark within the **BEIR-Fa** suite.

- **Language(s):** Persian (Farsi)  
- **Task(s):** Retrieval (Duplicate Question Retrieval)  
- **Source:** Translated from English using Google Translate  
- **Part of FaMTEB:** Yes — under BEIR-Fa

## Supported Tasks and Leaderboards

This dataset is designed to test **text embedding models** for their ability to retrieve **semantically similar or duplicate questions** within the domain of **Mathematica software and symbolic computation**. Results can be evaluated on the **Persian MTEB Leaderboard** (filter by language: Persian).

## Construction

The dataset was built by:

- Translating the "mathematica" StackExchange subforum using the **Google Translate API**
- Preserving the structure of community QA for **duplicate retrieval** evaluation

Translation quality, as detailed in the FaMTEB paper, was assessed by:

- **BM25 performance comparisons** with the original English versions  
- **GEMBA-DA framework** using LLMs for qualitative analysis

## Data Splits

According to the FaMTEB paper (Table 5), all CQADupstack-Fa datasets use the following combined test set:

- **Train:** 0 samples  
- **Dev:** 0 samples  
- **Test:** 480,902 samples (aggregate)

This particular sub-dataset includes approximately **18.9k examples** (user-provided). Specific train/dev/test splits for `cqadupstack-mathematica-fa` are not detailed in the FaMTEB paper. Refer to the dataset provider for finer-grained information.