farabi-lab/State_Session_Context_Passing

🇰🇿 Kazakh Stateful Session Context and Multi-Tool Agent Dataset

Dataset Summary

Kazakh Stateful Session Context and Multi-Tool Agent Dataset is a Kazakh-language dataset designed for training and evaluating Large Language Models (LLMs) in agentic AI workflows that require session context handling, multi-tool coordination, and state-aware response generation.

The dataset contains user requests, available tool schemas, expected tool calls, mock tool outputs, and full multi-turn interaction traces. It focuses on cases where an assistant must understand the user’s request, preserve relevant context, call multiple tools in sequence, and combine different information sources into a coherent final answer in Kazakh.

---

📊 Dataset Statistics

General Metrics

Metric	Count
Total Samples	3,280
Total Words (approx.)	918,829
Avg. Words per Sample	280

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	23.5	23.0	6	322	77,118
category	1.0	1.0	1	1	3,280
difficulty	1.0	1.0	1	1	3,280
id	1.0	1.0	1	1	3,280
query	12.9	12.0	3	41	42,226
tools	80.1	81.0	24	140	262,643
turns	160.7	157.0	62	689	527,002

---

Dataset Structure

Each dataset instance represents a complete state-aware tool-use interaction. A sample usually includes a Kazakh user query, available tools, expected tool calls, simulated tool outputs, and the final assistant answer.

Data Fields

• id: A unique identifier for the sample.

• query: The original user request in Kazakh. The query may require combining recent updates, background information, previous context, or multiple external sources.

• category: The task category. For example, 06_state_session_context indicates that the sample focuses on session state, context handling, and context-aware tool use.

• tools: A list of tools available to the assistant. Each tool contains:

  • name: the tool name;
  • description: the tool’s purpose;
  • parameters: the expected argument schema.

• difficulty: The difficulty level of the sample.

• answers: The target tool calls expected from the assistant. This field may include one or more tool calls that represent the intended action sequence.

• turns: The full interaction trajectory, including:

  • user request;
  • assistant planning step;
  • assistant tool calls;
  • mock tool responses;
  • final answer based on the retrieved information.

---

Data Instance

Below is one representative example from the dataset.

{
        "id": "sara_context_1",
        "category": "06_state_session_context",
        "query": "Назарбаев Университеті жайлы соңғы жаңалықтарды айтып бер және ол университет туралы мәлімет бер.",
        "tools": [
            {
                "name": "news.search",
                "description": "Search recent news articles",
                "parameters": {
                    "query": {
                        "type": "string",
                        "description": "Search topic",
                        "required": true
                    },
                    "language": {
                        "type": "string",
                        "description": "Language code",
                        "required": false
                    },
                    "pageToken": {
                        "type": "string",
                        "description": "Pagination token",
                        "required": false
                    }
                }
            },
            {
                "name": "wiki.search",
                "description": "Search Wikipedia articles",
                "parameters": {
                    "query": {
                        "type": "string",
                        "description": "Search term",
                        "required": true
                    },
                    "language": {
                        "type": "string",
                        "description": "Language code like kk, ru, en",
                        "required": false
                    }
                }
            }
        ],
        "difficulty": "hard",
        "answers": [
            {
                "name": "news.search",
                "arguments": "{\"query\": \"Назарбаев Университеті\", \"language\": \"kk\", \"pageToken\": \"??\"}"
            },
            {
                "name": "wiki.search",
                "arguments": "{\"query\": \"Назарбаев Университеті\", \"language\": \"kk\"}"
            }
        ],
        "turns": [
            {
                "role": "user",
                "content": "Назарбаев Университеті жайлы соңғы жаңалықтарды айтып бер және ол университет туралы мәлімет бер."
            },
            {
                "role": "assistant",
                "content": "жаңалықтан іздеу → wiki-ден іздеу деректерін өндеу",
                "meta": {
                    "plan": "news.search → wiki.search",
                    "tags": [],
                    "difficulty": "hard"
                }
            },
            {
                "role": "assistant",
                "content": "",
                "tool_call": {
                    "name": "news.search",
                    "arguments": "{\"query\": \"Назарбаев Университеті\", \"language\": \"kk\", \"pageToken\": \"??\"}"
                }
            },
            {
                "role": "tool",
                "content": "{\"articles\": [{\"source\": \"informburo.kz\", \"title\": \"Назарбаев Университетінде жас ғалымдар форумы өтті\", \"date\": \"2024-06-10\", \"url\": \"https://informburo.kz/kaz/new-nazarbayev-university-forum\"}]}"
            },
            {
                "role": "assistant",
                "content": "",
                "tool_call": {
                    "name": "wiki.search",
                    "arguments": "{\"query\": \"Назарбаев Университеті\", \"language\": \"kk\"}"
                }
            },
            {
                "role": "tool",
                "content": "{\"title\": \"Назарбаев Университеті\", \"summary\": \"Назарбаев Университеті — Қазақстан Республикасының жетекші зерттеу университеті, Астанада орналасқан, 2010 жылы құрылған.\", \"page_url\": \"https://kk.wikipedia.org/wiki/Назарбаев_Университеті\"}"
            },
            {
                "role": "assistant",
                "content": "Соңғы жаңалықтар бойынша, Назарбаев Университетінде жас ғалымдар форумы өтті (2024-06-10). Бұл туралы толығырақ informburo.kz сайтынан оқи аласыз: https://informburo.kz/kaz/new-nazarbayev-university-forum  \n    Назарбаев Университеті — Қазақстандағы жетекші зерттеу университеті, 2010 жылы Астанада құрылған. Университет жайлы қосымша ақпаратты мына сілтемеден қарауға болады: https://kk.wikipedia.org/wiki/Назарбаев_Университеті\n   (Дерек көзі: news.search + wiki.search)"
            }
        ]
    }

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

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}
}