hreyulog/multilingual-e5-small-arkts

SentenceTransformer based on intfloat/multilingual-e5-small

This is arkts model for Edge device.

This is a sentence-transformers model finetuned from intfloat/multilingual-e5-small. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval. docstring <--> passage:\npath: ...\nidentifier: ...\ncode: ...

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: intfloat/multilingual-e5-small
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'mean', 'include_prompt': True})
  (2): Normalize({})
)

Evaluation Results

On arkts-code-docstring dataset split test

Model	Params	MRR	NDCG@5	Recall@1	Recall@5
multilingual-e5-small-arkts	117.7M	0.6849	0.7078	0.6030	0.7952

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 = [
    'query: Persistent geographical location, re-enter to determine if the switch is turned on',
    'passage:\npath: code/BasicFeature/Media/Camera/entry/src/main/ets/Dialog/SettingDialog.ets\nidentifier: getLocationBol\ncode: getLocationBol(bol: boolean) {\n    this.settingDataObj.locationBol = bol;\n  }',
    'passage:\npath: custom_dialog/src/main/ets/model/modifier/TextAreaInputFilterModifier.ets\nidentifier: \ncode: export class TextAreaInputFilterModifier implements AttributeModifier<TextAreaAttribute> {\n  inputFilter?: InputFilter;\n\n  applyNormalAttribute(instance: TextAreaAttribute): void {\n    if (this.inputFilter) {\n      instance.inputFilter(this.inputFilter.value, this.inputFilter.error);\n    }\n  }\n}',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.4037, 0.0834],
#         [0.4037, 1.0000, 0.0600],
#         [0.0834, 0.0600, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 19,561 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1
  type	string	string
  details	min: 7 tokens mean: 31.93 tokens max: 256 tokens	min: 34 tokens mean: 157.48 tokens max: 256 tokens

• Samples:

sentence_0	sentence_1
query: 通过picker拉起图库并选择图片,并调用图片识码 @param options @returns	passage: path: libs/core/src/main/ets/utils/ScanUtils.ets identifier: onPickerScanForResult code: static async onPickerScanForResult(options?: scanBarcode.ScanOptions): Promise<Array<scanBarcode.ScanResult>> {   try {     let photoOption = new picker.PhotoSelectOptions()     photoOption.MIMEType = picker.PhotoViewMIMETypes.IMAGE_TYPE     photoOption.maxSelectNumber = 1     let photoPicker = new picker.PhotoViewPicker()     let uris = await photoPicker.select(photoOption)     return await ScanUtils.onDetectBarCode(uris[0], options)   } catch (err) {     let error = err as BusinessError;     LogUtils.debug('ScanUtils-onPickerScanForResult err', code: ${error.code} -·- message: ${error.message})     return [];   } }
query: 启动对话流程	passage: path: entry/src/main/ets/services/ai/ChatbotEngine.ets identifier: startConversation Flowcode: async startConversationFlow(   flowId: string,   sessionId: string,   initialContext: Record<string, any> = {} ): Promise<ConversationFlow | null> {   try {     const flowTemplate = this.getFlowTemplate(flowId);     if (!flowTemplate) {       throw new Error(Flow template not found: ${flowId});     }     const flow: ConversationFlow = {       id: ${flowId}_${Date.now()},       name: flowTemplate.name,       description: flowTemplate.description,       steps: [...flowTemplate.steps],       currentStep: 0,       isActive: true,       context: { ...initialContext }     };     this.activeFlows.set(sessionId, flow);     hilog.info(LogConstants.DOMAIN_APP, LogConstants.TAG_APP,       Started conversation flow: ${flowId} for session: ${sessionId});     return flow;   } catch (error) {     hilog.error(LogConstants.DOMAI...
query: 文本颜色接口	passage: path: entry/src/main/ets/common/types/CommonTypes.ets identifier: code: export interface TextColorConfig {   primary: string;   secondary: string;   disabled: string; }

{
    "scale": 20.0,
    "similarity_fct": "cos_sim",
    "gather_across_devices": false,
    "directions": [
        "query_to_doc"
    ],
    "partition_mode": "joint",
    "hardness_mode": null,
    "hardness_strength": 0.0
}

Training Hyperparameters

Non-Default Hyperparameters

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

All Hyperparameters

Click to expand

• per_device_train_batch_size: 32
• num_train_epochs: 1
• max_steps: -1
• learning_rate: 5e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0
• optim: adamw_torch_fused
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1
• label_smoothing_factor: 0.0
• bf16: False
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: False
• project: huggingface
• trackio_space_id: trackio
• per_device_eval_batch_size: 32
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: False
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: False
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• warmup_ratio: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.8170	500	0.6130

Training Time

• Training: 2.6 hours

Framework Versions

• Python: 3.14.3
• Sentence Transformers: 5.4.1
• Transformers: 5.5.4
• PyTorch: 2.11.0+cpu
• Accelerate: 1.13.0
• Datasets: 4.8.4
• Tokenizers: 0.22.2

Citation

BibTeX

ArkTS-CodeSearch

@misc{he2026arktscodesearchopensourcearktsdataset,
      title={ArkTS-CodeSearch: A Open-Source ArkTS Dataset for Code Retrieval}, 
      author={Yulong He and Artem Ermakov and Sergey Kovalchuk and Artem Aliev and Dmitry Shalymov},
      year={2026},
      eprint={2602.05550},
      archivePrefix={arXiv},
      primaryClass={cs.SE},
      url={https://arxiv.org/abs/2602.05550}, 
}