Leejy0-0/sbert-korean-triplet-mnr-v1

🔍 Model Overview

This model is a Korean Sentence-BERT fine-tuned with Triplet and MultipleNegativesRankingLoss. It is optimized for semantic similarity search in Korean construction accident reports.

• —Architecture: SBERT (Korean BERT backbone)
• —Training: Triplet + MNR
• —Language: Korean
• —Use cases: semantic search, STS, dense retrieval

SentenceTransformer

This is a sentence-transformers model trained. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

• —Model Type: Sentence Transformer <!-- - Base model: Unknown -->
• —Maximum Sequence Length: 512 tokens
• —Output Dimensionality: 768 dimensions
• —Similarity Function: Cosine Similarity <!-- - Training Dataset: Unknown --> <!-- - Language: Unknown --> <!-- - License: Unknown -->

Model Sources

• —Documentation: Sentence Transformers Documentation
• —Repository: Sentence Transformers on GitHub
• —Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    '21년 12월 11일(토) 17시 20분경, B/H 운전원 김**가 작업완료 후 운전석에서 내려오던 중 발이 미끄러지면서 복공판 상부로 넘어짐 (높이 1m) B/H 운전석 이탈 시 이동방법의 불량으로 인해 복공판 상부로 넘어짐운전석 발판을 통한 하차가 아닌 블레이드 경사판을 이용',
    '1층 실내 보 인테리어필름 작업 중, 근로자가 우마 다리 체결 상태를 정확하게 확인하지 않은 상태에서 우마 위에서 작업을 하다가 우마 다리가 접혀 넘어지면서 옆에 있던 우마 모서리에 코 부분을 부딪히는 사고 우마다리로 인한 넘어지면서 모서리에 코를 부딪힘',
    '2022년 05월 28일(토요일) 14시경 울산광역시 동구 일산동에 소재한 “스타벅스 울산 일산비치 D/T 신축공사” 현장에서 당사 직영 일용근로자인 최**(재해자)가 이동식 A형 사다리를 이용하여 1층 벽체 상부 타이 핀 제거 작업을 진행하는 과정에서 중심을 잃고 사다리와 함께 1.2M 정도의 높이에서 넘어져 우측 팔 부분에재해를 입는 사고가 발행하게 되었습니다. 사고 발생 후 울산대학교병원을 방문하여 진료 및 치료를 받았습니다. 이동식사다리를 이용한 작업중 넘어짐',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

<!--

Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details> -->

<!--

Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details> -->

<!--

Out-of-Scope Use

List how the model may foreseeably be misused and address what users ought not to do with the model. -->

<!--

Bias, Risks and Limitations

What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model. -->

<!--

Recommendations

What are recommendations with respect to the foreseeable issues? For example, filtering explicit content. -->

Training Details

Training Dataset

Unnamed Dataset

• —Size: 28,003 training samples
• —Columns: <code>sentence0</code> and <code>sentence1</code>
• —Approximate statistics based on the first 1000 samples: | | sentence0 | sentence1 | |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------| | type | string | string | | details | <ul><li>min: 8 tokens</li><li>mean: 51.87 tokens</li><li>max: 334 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 54.17 tokens</li><li>max: 325 tokens</li></ul> |
• —Samples: | sentence0 | sentence1 | |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | <code>철근 가공작업중 철근을 가공대에 올리는 과정에서 2인1조 작업중 1인이 선행 철근을 내려놓음으로 인하여 오른손 약지 손가락이 철근과 지면에 부딪힘 발생 철근 가공작업 중 철근을 가공대에 올리는 과정에서 2인1조 작업중 1인이 선행 철근을 내려놓음으로 인하여 오른손 약지 손가락에 철근에 부딪힘 발생</code> | <code>철근 절단 작업 중 고속절단기에 손가락 베임 작업자의 작업부주위에 의한 손가락 베임사고 발생</code> | | <code>전일 타설 후 슬라브 먹매김 작업을 위해 허리를 굽혀 통로 인근에서 작업중 슬라브 양생을 위한 살수로 바닥이 미끄러워 균형을 잃고 넘어지며 수직 철근에 우측 턱에 열상 발생 1. 이동 중 주의력 부족2. 콘크리트 양생을 위해 지면 살수로 인한 미끄럼 현상</code> | <code>지하1층 시스템동바리 가새 해체작업후 옆으로 이동중(높이2.8m) 지면으로 떨어짐 시스템 동바리의 가새 해체 하고 이동 중 안전고리 미체결로 떨어짐</code> | | <code>철근 절단기 상부 작업 중 발을 헛디디면서 절단기계에 가슴을 부딪히며 갈비뼈 골절 절단기 상부에 자재를 내리기 위해 한 발을 단에 올리고 한발로 지지하던 중, 자재를 내리는 동작 중 발이 미끄러져 헛디디며 공구에 가슴을 부딪힘</code> | <code>아침 TBM 조회 후 지상2층 외부 주차장에서 철근가공 작업을 하기 위하여, 가공작업 중 부주의로 인하여 (우측 검지) 손가락을 절곡기계에 협착하여 사고가 발생함 아침 TBM 조회 후 지상2층 외부 주차장에서 철근가공 작업을 하기 위하여, 가공작업 중 부주의로 인하여 (우측 검지) 손가락을 절곡기계에 협착하여 사고가 발생함</code> |
• —Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }

Training Hyperparameters

Non-Default Hyperparameters

• —per_device_train_batch_size: 32
• —per_device_eval_batch_size: 32
• —num_train_epochs: 1
• —multi_dataset_batch_sampler: round_robin

All Hyperparameters

<details><summary>Click to expand</summary>

• —overwrite_output_dir: False
• —do_predict: False
• —eval_strategy: no
• —prediction_loss_only: True
• —per_device_train_batch_size: 32
• —per_device_eval_batch_size: 32
• —per_gpu_train_batch_size: None
• —per_gpu_eval_batch_size: None
• —gradient_accumulation_steps: 1
• —eval_accumulation_steps: None
• —torch_empty_cache_steps: None
• —learning_rate: 5e-05
• —weight_decay: 0.0
• —adam_beta1: 0.9
• —adam_beta2: 0.999
• —adam_epsilon: 1e-08
• —max_grad_norm: 1
• —num_train_epochs: 1
• —max_steps: -1
• —lr_scheduler_type: linear
• —lr_scheduler_kwargs: {}
• —warmup_ratio: 0.0
• —warmup_steps: 0
• —log_level: passive
• —log_level_replica: warning
• —log_on_each_node: True
• —logging_nan_inf_filter: True
• —save_safetensors: True
• —save_on_each_node: False
• —save_only_model: False
• —restore_callback_states_from_checkpoint: False
• —no_cuda: False
• —use_cpu: False
• —use_mps_device: False
• —seed: 42
• —data_seed: None
• —jit_mode_eval: False
• —use_ipex: False
• —bf16: False
• —fp16: False
• —fp16_opt_level: O1
• —half_precision_backend: auto
• —bf16_full_eval: False
• —fp16_full_eval: False
• —tf32: None
• —local_rank: 0
• —ddp_backend: None
• —tpu_num_cores: None
• —tpu_metrics_debug: False
• —debug: []
• —dataloader_drop_last: False
• —dataloader_num_workers: 0
• —dataloader_prefetch_factor: None
• —past_index: -1
• —disable_tqdm: False
• —remove_unused_columns: True
• —label_names: None
• —load_best_model_at_end: False
• —ignore_data_skip: False
• —fsdp: []
• —fsdp_min_num_params: 0
• —fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
• —tp_size: 0
• —fsdp_transformer_layer_cls_to_wrap: None
• —accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
• —deepspeed: None
• —label_smoothing_factor: 0.0
• —optim: adamw_torch
• —optim_args: None
• —adafactor: False
• —group_by_length: False
• —length_column_name: length
• —ddp_find_unused_parameters: None
• —ddp_bucket_cap_mb: None
• —ddp_broadcast_buffers: False
• —dataloader_pin_memory: True
• —dataloader_persistent_workers: False
• —skip_memory_metrics: True
• —use_legacy_prediction_loop: False
• —push_to_hub: False
• —resume_from_checkpoint: None
• —hub_model_id: None
• —hub_strategy: every_save
• —hub_private_repo: None
• —hub_always_push: False
• —gradient_checkpointing: False
• —gradient_checkpointing_kwargs: None
• —include_inputs_for_metrics: False
• —include_for_metrics: []
• —eval_do_concat_batches: True
• —fp16_backend: auto
• —push_to_hub_model_id: None
• —push_to_hub_organization: None
• —mp_parameters:
• —auto_find_batch_size: False
• —full_determinism: False
• —torchdynamo: None
• —ray_scope: last
• —ddp_timeout: 1800
• —torch_compile: False
• —torch_compile_backend: None
• —torch_compile_mode: None
• —include_tokens_per_second: False
• —include_num_input_tokens_seen: False
• —neftune_noise_alpha: None
• —optim_target_modules: None
• —batch_eval_metrics: False
• —eval_on_start: False
• —use_liger_kernel: False
• —eval_use_gather_object: False
• —average_tokens_across_devices: False
• —prompts: None
• —batch_sampler: batch_sampler
• —multi_dataset_batch_sampler: round_robin

</details>

Training Logs

Framework Versions

• —Python: 3.11.12
• —Sentence Transformers: 3.4.1
• —Transformers: 4.51.3
• —PyTorch: 2.6.0+cu124
• —Accelerate: 1.6.0
• —Datasets: 3.5.1
• —Tokenizers: 0.21.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

<!--

Glossary

Clearly define terms in order to be accessible across audiences. -->

<!--

Model Card Authors

Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction. -->

<!--

Model Card Contact

Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors. -->