Qwen3-4B BKEE Entity Extraction

A Vietnamese entity extraction model fine-tuned from unsloth/Qwen3-4B-Instruct-2507. Given a sentence, the model identifies candidate entity mentions (Person, Organization, Job, etc.) along with a step-by-step natural-language reasoning trace explaining why each span was selected.

This is one of four task-specific checkpoints from a distant-supervision Vietnamese event-extraction pipeline, alongside:

• hadeseus/qwen3-4b-bkee-event-extraction
• hadeseus/qwen3-4b-bkee-argument-assigner
• hadeseus/qwen3-4b-bkee-full-extraction

Training Data

The model was trained on a silver-labeled reasoning dataset derived from BKEE (a Vietnamese Event Extraction dataset introduced at LREC-COLING 2024, containing 1,066 manually annotated documents across 33+ event types and 28 argument roles).

The silver data was produced via a distant supervision pipeline:

1. Candidate entity spans and BKEE gold labels are used to prompt an LLM to generate step-by-step Chain-of-Thought reasoning explaining the correct entity type for each span.
2. Predictions are validated against Pydantic schemas and checked against BKEE gold annotations (non-empty gold labels are strictly matched; sentences with empty gold labels allow the LLM to surface additional plausible entities).
3. Verified reasoning traces are combined with the original BKEE sentences to form the supervised fine-tuning dataset for this checkpoint.

Note: this model is trained on machine-generated (silver) reasoning data aligned to BKEE's gold entity labels, not on human-written explanations.

Intended Use

Given a Vietnamese sentence, the model outputs a reasoning trace followed by the extracted entity mentions and their types. It is intended as the entity-extraction stage of a larger event extraction pipeline (see the full-extraction checkpoint for an end-to-end variant).

Usage

System prompt:

Trích xuất các đề cập thực thể từ câu tiếng Việt đã cho. Hãy suy nghĩ theo từng bước.

User prompt: the raw sentence. Output: reasoning followed by {"entities": [{"text": ..., "type": ...}]}.

Citation

If you use this model, please cite the underlying BKEE dataset:

@inproceedings{nguyen-etal-2024-bkee,
    title = "{BKEE}: Pioneering Event Extraction in the {V}ietnamese Language",
    author = "Nguyen, Thi-Nhung  and
      Tran, Bang Tien  and
      Luu, Trong-Nghia  and
      Nguyen, Thien Huu  and
      Nguyen, Kiem-Hieu",
    booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
    month = may,
    year = "2024",
    address = "Torino, Italia",
    publisher = "ELRA and ICCL",
    url = "https://aclanthology.org/2024.lrec-main.217",
    pages = "2421--2427"
}