| --- |
| language: |
| - kk |
| license: cc-by-nc-4.0 |
| task_categories: |
| - text-generation |
| - question-answering |
| tags: |
| - kazakh |
| - agentic-ai |
| - tool-use |
| - function-calling |
| - tool-output-interpretation |
| - financial-assistant |
| - banking-tools |
| - multi-tool-use |
| - structured-output |
| - llm-agents |
| pretty_name: Kazakh Tool Output Interpretation Dataset |
| size_categories: |
| - 1K<n<10K |
| --- |
| |
| # 🇰🇿 Kazakh Tool Output Interpretation and Financial Action Dataset |
|
|
| ## Dataset Summary |
|
|
| **Kazakh Tool Output Interpretation and Financial Action Dataset** is a Kazakh-language dataset designed for training and evaluating Large Language Models (LLMs) in tool-augmented agentic workflows that require interpreting structured tool outputs and generating grounded final responses. |
|
|
| The dataset focuses on scenarios where an assistant must understand a Kazakh user request, call the appropriate tools, read structured outputs, and summarize the result accurately. Many samples involve sensitive or action-oriented domains, such as banking, account balance checking, transfers, and financial status reporting. |
|
|
| This dataset is useful for studying how Kazakh-language AI agents interpret tool responses, combine outputs from multiple tools, preserve numerical accuracy, and produce concise user-facing answers based on simulated external APIs. |
|
|
| --- |
|
|
| ## 📊 Dataset Statistics |
|
|
| ### General Metrics |
|
|
| | Metric | Count | |
| | :--- | :--- | |
| | **Total Samples** | 3,568 | |
| | **Total Words** (approx.) | 1,079,862 | |
| | **Avg. Words per Sample** | 302 | |
|
|
| ### Word Count Distribution Per Field |
|
|
| The following table details the distribution of word counts across different fields in the dataset. |
|
|
| | Field | Mean | Median | Min | Max | Total Words | |
| | :--- | :--- | :--- | :--- | :--- | :--- | |
| | **answers** | 29.3 | 26.0 | 8 | 137 | 104,465 | |
| | **category** | 1.0 | 1.0 | 1 | 1 | 3,568 | |
| | **difficulty** | 1.0 | 1.0 | 1 | 1 | 3,568 | |
| | **id** | 1.0 | 1.0 | 1 | 1 | 3,568 | |
| | **query** | 15.3 | 15.0 | 3 | 38 | 54,731 | |
| | **tools** | 88.7 | 86.0 | 30 | 160 | 316,565 | |
| | **turns** | 166.3 | 160.0 | 66 | 369 | 593,397 | |
|
|
|
|
|  |
|
|
| --- |
|
|
| ## Dataset Structure |
|
|
| Each dataset instance represents a complete tool-use interaction. A sample usually includes a Kazakh user query, available tool schemas, expected tool calls, simulated tool responses, and a final assistant answer. |
|
|
| ### Data Fields |
|
|
| - **`id`**: A unique identifier for the sample. |
|
|
| - **`query`**: The original user request in Kazakh. The query may involve checking information, performing a simulated action, or asking the assistant to interpret structured tool results. |
|
|
| - **`category`**: The task category. For example, `07_tool_output_interpretation` indicates that the sample focuses on understanding and summarizing tool outputs correctly. |
|
|
| - **`tools`**: A list of tools available to the assistant. Each tool contains: |
| - `name`: the tool name; |
| - `description`: the tool’s purpose; |
| - `parameters`: the required or optional input schema; |
| - `mock_response`: the expected response structure. |
|
|
| - **`difficulty`**: The difficulty level of the sample. |
|
|
| - **`answers`**: The expected tool calls. This field may contain one or more function calls with serialized JSON arguments. |
|
|
| - **`turns`**: The complete interaction trace, including: |
| - user request; |
| - assistant planning; |
| - assistant tool calls; |
| - mock tool responses; |
| - final answer grounded in the tool outputs. |
|
|
| --- |
|
|
| ## Data Instance |
|
|
| Below is one representative example from the dataset. |
|
|
| ```json |
| { |
| "id": "dastan_07_1", |
| "query": "Менің 123456789 шотымдағы ақшаны 987654321 шотына 10000 теңге аударшы.", |
| "category": "07_tool_output_interpretation", |
| "tools": [ |
| { |
| "name": "bank.balance", |
| "description": "Get bank account balance", |
| "parameters": { |
| "account": { |
| "type": "string", |
| "description": "Account number", |
| "required": true |
| }, |
| "api_key": { |
| "type": "string", |
| "description": "Auth key", |
| "required": false |
| } |
| }, |
| "mock_response": { |
| "account": "", |
| "balance": null, |
| "currency": "" |
| } |
| }, |
| { |
| "name": "bank.transfer", |
| "description": "Transfer money between accounts", |
| "parameters": { |
| "from_account": { |
| "type": "string", |
| "description": "Source account", |
| "required": true |
| }, |
| "to_account": { |
| "type": "string", |
| "description": "Destination account", |
| "required": true |
| }, |
| "amount": { |
| "type": "float", |
| "description": "Amount to transfer", |
| "required": true |
| }, |
| "api_key": { |
| "type": "string", |
| "description": "Auth key", |
| "required": true |
| } |
| }, |
| "mock_response": { |
| "transaction_id": "", |
| "status": "", |
| "amount": null, |
| "currency": "", |
| "remaining_balance": null |
| } |
| } |
| ], |
| "difficulty": "easy", |
| "answers": [ |
| { |
| "name": "bank.balance", |
| "arguments": "{\"account\": \"123456789\", \"api_key\": \"example-key\"}" |
| }, |
| { |
| "name": "bank.transfer", |
| "arguments": "{\"from_account\": \"123456789\", \"to_account\": \"987654321\", \"amount\": 10000.0, \"api_key\": \"example-key\"}" |
| } |
| ], |
| "turns": [ |
| { |
| "role": "user", |
| "content": "Менің 123456789 шотымдағы ақшаны 987654321 шотына 10000 теңге аударшы." |
| }, |
| { |
| "role": "assistant", |
| "content": "Алдымен балансты тексеріп, сосын аударым жасаймын.", |
| "meta": { |
| "plan": "Extract and normalize bank.balance + bank.transfer data", |
| "tags": [], |
| "difficulty": "hard" |
| } |
| }, |
| { |
| "role": "assistant", |
| "content": "", |
| "tool_call": { |
| "name": "bank.balance", |
| "arguments": "{\"account\": \"123456789\", \"api_key\": \"example-key\"}" |
| } |
| }, |
| { |
| "role": "tool", |
| "content": "{\"account\": \"123456789\", \"balance\": 25000.0, \"currency\": \"KZT\"}" |
| }, |
| { |
| "role": "assistant", |
| "content": "", |
| "tool_call": { |
| "name": "bank.transfer", |
| "arguments": "{\"from_account\": \"123456789\", \"to_account\": \"987654321\", \"amount\": 10000.0, \"api_key\": \"example-key\"}" |
| } |
| }, |
| { |
| "role": "tool", |
| "content": "{\"transaction_id\": \"tr12345\", \"status\": \"success\", \"amount\": 10000.0, \"currency\": \"KZT\", \"remaining_balance\": 15000.0}" |
| }, |
| { |
| "role": "assistant", |
| "content": "123456789 шотынан 987654321 шотына 10000 теңге сәтті аударылды. Қалған баланс: 15000 теңге. (Дерек көзі: bank.balance + bank.transfer)" |
| } |
| ] |
| }, |
| ``` |
|
|
| ## Funding |
|
|
| This dataset was developed as part of the project funded by the Ministry of Science and Higher Education of the Republic of Kazakhstan under Grant No. BR24993001, “Creation of a Large Language Model (LLM) to Support the Kazakh Language and Advance Technological Development.” |
|
|
| ## Citation |
|
|
| If you use this dataset in your research, please cite the following article: |
|
|
| ### APA |
|
|
| Kadyrbek, N., Tuimebayev, Z., Mansurova, M., & Viegas, V. (2025). The development of small-scale language models for low-resource languages, with a focus on Kazakh and direct preference optimization. *Big Data and Cognitive Computing, 9*(5), 137. [https://doi.org/10.3390/bdcc9050137](https://doi.org/10.3390/bdcc9050137) |
|
|
| ### BibTeX |
|
|
| ```bibtex |
| @article{kadyrbek2025development, |
| title = {The Development of Small-Scale Language Models for Low-Resource Languages, with a Focus on Kazakh and Direct Preference Optimization}, |
| author = {Kadyrbek, Nurgali and Tuimebayev, Zhanseit and Mansurova, Madina and Viegas, Vitor}, |
| journal = {Big Data and Cognitive Computing}, |
| volume = {9}, |
| number = {5}, |
| pages = {137}, |
| year = {2025}, |
| publisher = {MDPI}, |
| doi = {10.3390/bdcc9050137}, |
| url = {https://www.mdpi.com/2504-2289/9/5/137} |
| } |
| ``` |
|
|
