CoolFace
Modelpublic

GyunYeop/midm-base-GRPO-lora-tuning-KoreanCultureQA

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes10downloads
Model Card

Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

Model Details

본 모델은 국립국어원 주최 2025년 인공지능의 한국어 능력 평가 경진대회 [2025]한국문화 질의응답(가 유형) ISNLP팀 최종 제출물이다.

Model Description

<!-- Provide a longer summary of what this model is. -->

학습데이터는 2025년 인공지능의 한국어 능력 평가 경진대회 [2025]한국문화 질의응답(가 유형)에서 주어진 train dataset을 이용하여 학습하였다.

data link(현재 다운로드 불가): http://kli.korean.go.kr/taskOrdtm/taskList.do?taskOrdtmId=180&clCd=END_TASK&subMenuId=sub01

해당 과제에서는 Midm-base+QLoRA모델을 기반으로 GRPO + Descriptive Answer Candidate로 학습하였다.

자세한 학습 방법론 및 학습 코드는 https://github.com/KimGyunYeop/2025MalPyeongQAISNLPRLVR_WTA 에서 확인할 수 있다.

Model Usage

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, GenerationConfig
from peft import LoraConfig, PeftModelForCausalLM
import torch
    
adapter_path = "GyunYeop/midm-base-GRPO-lora-tuning-KoreanCultureQA"
lora_config = LoraConfig.from_pretrained(adapter_path)
model_name = lora_config.base_model_name_or_path
tokenizer = AutoTokenizer.from_pretrained(model_name)
generation_config= GenerationConfig.from_pretrained(model_name)

bnb_config = BitsAndBytesConfig(
        load_in_4bit=True,                       # 4‑bit 가중치
        bnb_4bit_quant_type="nf4",               # Normal‑Float 4
        bnb_4bit_use_double_quant=True,          # double‑quant
        bnb_4bit_compute_dtype=torch.bfloat16,   # Ada, Hopper, MI300 등
        llm_int8_skip_modules=["lm_head"]        # 출력층은 FP16
    )
    
base_model = AutoModelForCausalLM.from_pretrained(model_name, 
                                                device_map="cuda:0",
                                                trust_remote_code=True,
                                                quantization_config=bnb_config,
                                                )

model = PeftModelForCausalLM.from_pretrained(base_model, adapter_path)
model.load_adapter(adapter_path, subfolder="selection", adapter_name="selection")

prompt = '단답형 문제에서 정답을 맞추기위해 반드시 충분히 생각해보고 "생각 과정:" 이후에 정답 근거 및 생각과정을 작성 한 다음 이후 최종 답변을 생성할 것 (문제 유형: 단답형 생각과정: {생각과정} 답변: {최종답변} 형태)'
question = '문제 유형: 단답형 \n 질문: 2005년에 개관하였으며, 교육 및 문화적 목적으로 영화를 상영하는 서울의 유일한 비영리 민간 시네마테크 전용관은 어디인가요?'
inputs = tokenizer.apply_chat_template(
    [
        {"role": "system", "content": prompt},
        {"role": "user", "content": question}
    ],
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt"
)

answer = model.generate(
    inputs.to("cuda:0"),
    generation_config=generation_config,
    max_new_tokens=1024,
    do_sample=True
)

answer_text = tokenizer.decode(answer[0][inputs.shape[-1]:], skip_special_tokens=True)
print(answer_text)

<!--

  • —Developed by: [More Information Needed]
  • —Funded by [optional]: [More Information Needed]
  • —Shared by [optional]: [More Information Needed]
  • —Model type: [More Information Needed]
  • —Language(s) (NLP): [More Information Needed]
  • —License: [More Information Needed]
  • —Finetuned from model [optional]: [More Information Needed]

Model Sources [optional]

<!-- Provide the basic links for the model. -->

  • —Repository: [More Information Needed]
  • —Paper [optional]: [More Information Needed]
  • —Demo [optional]: [More Information Needed]

Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->

[More Information Needed]

Downstream Use [optional]

<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->

[More Information Needed]

Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->

[More Information Needed]

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

[More Information Needed]

Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

Training Details

Training Data

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

[More Information Needed]

Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

Preprocessing [optional]

[More Information Needed]

Training Hyperparameters
  • —Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
Speeds, Sizes, Times [optional]

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

[More Information Needed]

Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

Testing Data, Factors & Metrics

Testing Data

<!-- This should link to a Dataset Card if possible. -->

[More Information Needed]

Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

[More Information Needed]

Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

[More Information Needed]

Results

[More Information Needed]

Summary

Model Examination [optional]

<!-- Relevant interpretability work for the model goes here -->

[More Information Needed]

Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

  • —Hardware Type: [More Information Needed]
  • —Hours used: [More Information Needed]
  • —Cloud Provider: [More Information Needed]
  • —Compute Region: [More Information Needed]
  • —Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]

Framework versions

  • —PEFT 0.16.0 -->