bge-reranker-v2-m3-fds-finetuned

CrossEncoder based on BAAI/bge-reranker-v2-m3

This is a Cross Encoder model finetuned from BAAI/bge-reranker-v2-m3 using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

• Model Type: Cross Encoder
• Base model: BAAI/bge-reranker-v2-m3 <!-- at revision 953dc6f6f85a1b2dbfca4c34a2796e7dde08d41e -->
• Maximum Sequence Length: 2048 tokens
• Number of Output Labels: 1 label

<!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

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 CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("cross_encoder_model_id")
# Get scores for pairs of texts
pairs = [
    ['What is the formula that returns the number of brokers making  a quarterly estimate for Cash flow from investing over last 6 years?', 'Formula: FF_INVEST_PURCH_SALE_CF\nName: Purchase/Sale of Investments\nKeywords: net cashflow on purchase of Investments; net cash for Purchase and Sale of Investments; cash outflow or inflow for investment; net Purchase and Sale of Investments cash flow; net cash generated from sales of investment\nDescription: This formula returns the net Purchase of Investments figures of a company on the cash flow statement.  It represents the sum of the negative of Purchase of Investments and Sale/Maturity of Investments\nExample Question: Can you give me the formula that returns the net Purchase of Investments of a company?\nFactSet Description: Annual, Interim, and Preliminary Items - All Industries\nReturns purchase/sale of investments for the period and date(s) requested in local currency by default.\nThis is calculated as the sum of the negative of Purchase of Investments (FF_INVEST_PURCH_CF) and Sale/Maturity of Investments (FF_INVEST_SALE_CF).\nIf Increase in Investments (FF_INVEST_PURCH_CF) or Decrease in Investments (FF_INVEST_SALE_CF) are not available, zero will be substituted.'],
    ['Give me the code that returns the annual assets values since 2017?', 'Formula: FF_ROEA\nName: Return on Earning Assets\nKeywords: Return on Earning Assets; Net Income After Extraordinary Items over average total Investments; income over investments\nDescription: This formula returns the Return on Earning Assets of a company.\nExample Question: Can you give me the formula that returns the Return on Earning Assets of a company?\nFactSet Description: Annual and Interim Items\nRepresents the return on earnings assets for the period and date(s) requested.\nThis is calculated as Net Income After Extraordinary Items (FF_NET_INC_AFT_XORD) divided by the sum of the two fiscal period average of Total Investments (FF_BK_INVEST_TOT) and the two fiscal period average of Net Loans (FF_LOAN_NET).\nThe result is then multiplied by 100.\nIf data is requested on a quarterly or semiannual frequency, Return on Earning Assets uses the LTM value for Net Income After Extraordinary Items (FF_NET_INC_AFT_XORD).'],
    ['you gave me 2 codes: =FDS(A1,"FF_GEOREV_MKT_PCT(""DEV"",ANN,0)") and =FDS(A1,"FF_GEOREV_MKT_PCT(""EMG"",ANN,0)"), they are supposed to get me all revenue generated from both developed and developing markets, how come it doesn\'t add up to 100%?', "Formula: FF_GEOREV_EX_DOMICILE_PCT\nName: FF_GEOREV_EX_DOMICILE_PCT\nKeywords: percent of revenue of a company generated outside a certain country; % of a company's revenue share excluding a certain country; percentage of a company's sales ex- specific country\nDescription: This formula returns a percentage data which gives the percentage of revenue generated outside of a specific country (specified in the parameter of this formula).  The data returned by FF_GEOREV_EX_DOMICILE_PCT is opposite to FF_GEOREV_COUNTRY_PCT ( which gives the percentage of sales generated from a specific country).  In other words: FF_GEOREV_EX_DOMICILE_PCT = 100 -  FF_GEOREV_COUNTRY_PCT.  Although FF_GEOREV_EX_DOMICILE_PCT allows time series capability, it does not work.  Treat this formula as a single point formula where you can specify a specific year of data.\nExample Question: Do you have a code that tells me the percentage of a company's sales generated outside a specific country?\nFactSet Description: "],
    ['You have enterprise value formula?', 'Formula: FF_ENTRPR_VAL_DAILY\nName: Enterprise Value - Daily\nKeywords: Enterprise Value; EV; EV daily update; daily EV; weekly EV; monthly EV\nDescription: This formula returns EV that updates daily according to the price.  Use this formula if user simply requests for EV or enterprise value.  This is a most common formula and allows users to get EV as of a specific date using specific date prices.  If user request for EV or enterprise value for a fiscal period end, then use FF_ENTRPR_VAL instead.  FF_ENTRPR_VAL_DAILY provides EV on daily basis, updated with each day closing price, suitable if user request for daily, weekly or monthly EV.  On the other hand, FF_ENTRPR_VAL provides enterprise value (EV) on fiscal end, such as annual, semi-annual or quarter, these figures do not update in between fiscal period ends, more suitable if users are requesting for fiscal period end figure.\nExample Question: Can you give me the formula that returns the Enterprise Value  of a company?\nFactSet Description: Returns Enterprise Value - Daily for the period and date(s) requested in local currency by default. This returns a parent company-level valuation.\nThis is calculated as the sum of:\nPrice (P_PRICE) multiplied by Shares;\nTotal Debt (FF_DEBT);\nPreferred Stock - Carrying Value (FF_PFD_STK);\nTotal Accumulated Minority Interest (FF_MIN_INT_ACCUM)\nminus Cash.\nFor Shares: If you select Diluted Shares, Shares is defined as Common Shares Used To Calculate Fully Diluted Earnings Per Share (FF_COM_SHS_OUT_EPS_DIL). If this is not available, then Common Shares Outstanding (FF_COM_SHS_OUT) will be substituted. If you select Shares Outstanding, Shares is defined as Common Shares Outstanding (FF_COM_SHS_OUT) or Common Shares Outstanding - Current (FF_COM_SHS_OUT_CURR), whichever is more recent. If Common Shares Outstanding (FF_COM_SHS_OUT) is not available, then Common Shares Used to Calculate Basic Earnings Per Share (FF_COM_SHS_OUT_EPS_BASIC) will be substituted.\nFor Cash: For companies in the Insurance industry, cash is defined as Cash Only (FF_CASH_ONLY). For all other companies, cash is defined as Cash & Short Term Investments (FF_CASH_ST). If FF_CASH_ST is not available, Cash & Due from Banks (FF_CASH_DUE_FR_BK) will be substituted.\nFor Price: Price is fetched from the identifier returned by FS_PARENT_EQUITY and not necessarily the identifier that is being used in the downloading formula.\nFor Debt: The default formula used is FF_DEBT which include operating lease. If you exclude operating lease, the formula called is FF_DEBT_XOPER.\nNote\tCalculation excludes pension liabilities.\nRestated report bases are used where available.'],
    ['What is the code that returns a net income estimate for a specific year as of a certain date', 'Formula: FF_NET_INC_DIL_AFT_XORD\nName: Net Income Avail to Common - Fully Diluted - After Extraordinaries\nKeywords:  diluted net income after extraordinaries; Fully Diluted Net Income Available to Common shareholders After Extraordinaries; net income available to common fully diluted EPS after extraordinaries; diluted net income available to common shareholder after extraordinaries;\nDescription: This formula returns the Fully Diluted Net Income Available to Common shareholders After Extraordinaries figures of a company on the income statement.  It represents net income available to common fully diluted EPS after extraordinaries\nExample Question: Can you give me the formula that returns the Fully Diluted Net Income Available to Common shareholders After Extraordinaries of a company?\nFactSet Description: Annual, Interim and Preliminary Items - All Industries\nRepresents net income available to common fully diluted EPS after extraordinaries for the period and date(s) requested in local currency by default.\nThis is calculated as the sum of Net Income Used to Calculate Diluted Earnings Per Share (FF_NET_INC_DIL_BEF_DISC_OPER) and Extraordinary Items & Gain/Loss on Sale of Assets (FF_XORD_ITEMS).\nZero is subsituted for Extraordinary Items & Gain/Loss on Sale of Assets (FF_XORD_ITEMS) if it is not available.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'What is the formula that returns the number of brokers making  a quarterly estimate for Cash flow from investing over last 6 years?',
    [
        'Formula: FF_INVEST_PURCH_SALE_CF\nName: Purchase/Sale of Investments\nKeywords: net cashflow on purchase of Investments; net cash for Purchase and Sale of Investments; cash outflow or inflow for investment; net Purchase and Sale of Investments cash flow; net cash generated from sales of investment\nDescription: This formula returns the net Purchase of Investments figures of a company on the cash flow statement.  It represents the sum of the negative of Purchase of Investments and Sale/Maturity of Investments\nExample Question: Can you give me the formula that returns the net Purchase of Investments of a company?\nFactSet Description: Annual, Interim, and Preliminary Items - All Industries\nReturns purchase/sale of investments for the period and date(s) requested in local currency by default.\nThis is calculated as the sum of the negative of Purchase of Investments (FF_INVEST_PURCH_CF) and Sale/Maturity of Investments (FF_INVEST_SALE_CF).\nIf Increase in Investments (FF_INVEST_PURCH_CF) or Decrease in Investments (FF_INVEST_SALE_CF) are not available, zero will be substituted.',
        'Formula: FF_ROEA\nName: Return on Earning Assets\nKeywords: Return on Earning Assets; Net Income After Extraordinary Items over average total Investments; income over investments\nDescription: This formula returns the Return on Earning Assets of a company.\nExample Question: Can you give me the formula that returns the Return on Earning Assets of a company?\nFactSet Description: Annual and Interim Items\nRepresents the return on earnings assets for the period and date(s) requested.\nThis is calculated as Net Income After Extraordinary Items (FF_NET_INC_AFT_XORD) divided by the sum of the two fiscal period average of Total Investments (FF_BK_INVEST_TOT) and the two fiscal period average of Net Loans (FF_LOAN_NET).\nThe result is then multiplied by 100.\nIf data is requested on a quarterly or semiannual frequency, Return on Earning Assets uses the LTM value for Net Income After Extraordinary Items (FF_NET_INC_AFT_XORD).',
        "Formula: FF_GEOREV_EX_DOMICILE_PCT\nName: FF_GEOREV_EX_DOMICILE_PCT\nKeywords: percent of revenue of a company generated outside a certain country; % of a company's revenue share excluding a certain country; percentage of a company's sales ex- specific country\nDescription: This formula returns a percentage data which gives the percentage of revenue generated outside of a specific country (specified in the parameter of this formula).  The data returned by FF_GEOREV_EX_DOMICILE_PCT is opposite to FF_GEOREV_COUNTRY_PCT ( which gives the percentage of sales generated from a specific country).  In other words: FF_GEOREV_EX_DOMICILE_PCT = 100 -  FF_GEOREV_COUNTRY_PCT.  Although FF_GEOREV_EX_DOMICILE_PCT allows time series capability, it does not work.  Treat this formula as a single point formula where you can specify a specific year of data.\nExample Question: Do you have a code that tells me the percentage of a company's sales generated outside a specific country?\nFactSet Description: ",
        'Formula: FF_ENTRPR_VAL_DAILY\nName: Enterprise Value - Daily\nKeywords: Enterprise Value; EV; EV daily update; daily EV; weekly EV; monthly EV\nDescription: This formula returns EV that updates daily according to the price.  Use this formula if user simply requests for EV or enterprise value.  This is a most common formula and allows users to get EV as of a specific date using specific date prices.  If user request for EV or enterprise value for a fiscal period end, then use FF_ENTRPR_VAL instead.  FF_ENTRPR_VAL_DAILY provides EV on daily basis, updated with each day closing price, suitable if user request for daily, weekly or monthly EV.  On the other hand, FF_ENTRPR_VAL provides enterprise value (EV) on fiscal end, such as annual, semi-annual or quarter, these figures do not update in between fiscal period ends, more suitable if users are requesting for fiscal period end figure.\nExample Question: Can you give me the formula that returns the Enterprise Value  of a company?\nFactSet Description: Returns Enterprise Value - Daily for the period and date(s) requested in local currency by default. This returns a parent company-level valuation.\nThis is calculated as the sum of:\nPrice (P_PRICE) multiplied by Shares;\nTotal Debt (FF_DEBT);\nPreferred Stock - Carrying Value (FF_PFD_STK);\nTotal Accumulated Minority Interest (FF_MIN_INT_ACCUM)\nminus Cash.\nFor Shares: If you select Diluted Shares, Shares is defined as Common Shares Used To Calculate Fully Diluted Earnings Per Share (FF_COM_SHS_OUT_EPS_DIL). If this is not available, then Common Shares Outstanding (FF_COM_SHS_OUT) will be substituted. If you select Shares Outstanding, Shares is defined as Common Shares Outstanding (FF_COM_SHS_OUT) or Common Shares Outstanding - Current (FF_COM_SHS_OUT_CURR), whichever is more recent. If Common Shares Outstanding (FF_COM_SHS_OUT) is not available, then Common Shares Used to Calculate Basic Earnings Per Share (FF_COM_SHS_OUT_EPS_BASIC) will be substituted.\nFor Cash: For companies in the Insurance industry, cash is defined as Cash Only (FF_CASH_ONLY). For all other companies, cash is defined as Cash & Short Term Investments (FF_CASH_ST). If FF_CASH_ST is not available, Cash & Due from Banks (FF_CASH_DUE_FR_BK) will be substituted.\nFor Price: Price is fetched from the identifier returned by FS_PARENT_EQUITY and not necessarily the identifier that is being used in the downloading formula.\nFor Debt: The default formula used is FF_DEBT which include operating lease. If you exclude operating lease, the formula called is FF_DEBT_XOPER.\nNote\tCalculation excludes pension liabilities.\nRestated report bases are used where available.',
        'Formula: FF_NET_INC_DIL_AFT_XORD\nName: Net Income Avail to Common - Fully Diluted - After Extraordinaries\nKeywords:  diluted net income after extraordinaries; Fully Diluted Net Income Available to Common shareholders After Extraordinaries; net income available to common fully diluted EPS after extraordinaries; diluted net income available to common shareholder after extraordinaries;\nDescription: This formula returns the Fully Diluted Net Income Available to Common shareholders After Extraordinaries figures of a company on the income statement.  It represents net income available to common fully diluted EPS after extraordinaries\nExample Question: Can you give me the formula that returns the Fully Diluted Net Income Available to Common shareholders After Extraordinaries of a company?\nFactSet Description: Annual, Interim and Preliminary Items - All Industries\nRepresents net income available to common fully diluted EPS after extraordinaries for the period and date(s) requested in local currency by default.\nThis is calculated as the sum of Net Income Used to Calculate Diluted Earnings Per Share (FF_NET_INC_DIL_BEF_DISC_OPER) and Extraordinary Items & Gain/Loss on Sale of Assets (FF_XORD_ITEMS).\nZero is subsituted for Extraordinary Items & Gain/Loss on Sale of Assets (FF_XORD_ITEMS) if it is not available.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

<!-- ### 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.* -->

Evaluation

Metrics

Cross Encoder Correlation

• Dataset: fds_bge_regression_eval
• Evaluated with <code>CECorrelationEvaluator</code>

<!-- ## 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: 8,378 training samples
• Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	<ul><li>min: 21 characters</li><li>mean: 96.28 characters</li><li>max: 331 characters</li></ul>	<ul><li>min: 23 characters</li><li>mean: 1569.56 characters</li><li>max: 5992 characters</li></ul>	<ul><li>min: 0.0</li><li>mean: 0.29</li><li>max: 1.0</li></ul>

• Samples:

  sentence_0	sentence_1	label
  <code>What is the formula that returns the number of brokers making a quarterly estimate for Cash flow from investing over last 6 years?</code>	<code>Formula: FF_INVEST_PURCH_SALE_CF<br>Name: Purchase/Sale of Investments<br>Keywords: net cashflow on purchase of Investments; net cash for Purchase and Sale of Investments; cash outflow or inflow for investment; net Purchase and Sale of Investments cash flow; net cash generated from sales of investment<br>Description: This formula returns the net Purchase of Investments figures of a company on the cash flow statement. It represents the sum of the negative of Purchase of Investments and Sale/Maturity of Investments<br>Example Question: Can you give me the formula that returns the net Purchase of Investments of a company?<br>FactSet Description: Annual, Interim, and Preliminary Items - All Industries<br>Returns purchase/sale of investments for the period and date(s) requested in local currency by default.<br>This is calculated as the sum of the negative of Purchase of Investments (FF_INVEST_PURCH_CF) and Sale/Maturity of Investments (FF_INVEST_SALE_CF).<br>If Increase in Investments (FF_INVEST_PURCH_CF) or Decr...</code>	<code>0.0</code>
  <code>Give me the code that returns the annual assets values since 2017?</code>	<code>Formula: FF_ROEA<br>Name: Return on Earning Assets<br>Keywords: Return on Earning Assets; Net Income After Extraordinary Items over average total Investments; income over investments<br>Description: This formula returns the Return on Earning Assets of a company.<br>Example Question: Can you give me the formula that returns the Return on Earning Assets of a company?<br>FactSet Description: Annual and Interim Items<br>Represents the return on earnings assets for the period and date(s) requested.<br>This is calculated as Net Income After Extraordinary Items (FF_NET_INC_AFT_XORD) divided by the sum of the two fiscal period average of Total Investments (FF_BK_INVEST_TOT) and the two fiscal period average of Net Loans (FF_LOAN_NET).<br>The result is then multiplied by 100.<br>If data is requested on a quarterly or semiannual frequency, Return on Earning Assets uses the LTM value for Net Income After Extraordinary Items (FF_NET_INC_AFT_XORD).</code>	<code>0.0</code>
  <code>you gave me 2 codes: =FDS(A1,"FF_GEOREV_MKT_PCT(""DEV"",ANN,0)") and =FDS(A1,"FF_GEOREV_MKT_PCT(""EMG"",ANN,0)"), they are supposed to get me all revenue generated from both developed and developing markets, how come it doesn't add up to 100%?</code>	<code>Formula: FF_GEOREV_EX_DOMICILE_PCT<br>Name: FF_GEOREV_EX_DOMICILE_PCT<br>Keywords: percent of revenue of a company generated outside a certain country; % of a company's revenue share excluding a certain country; percentage of a company's sales ex- specific country<br>Description: This formula returns a percentage data which gives the percentage of revenue generated outside of a specific country (specified in the parameter of this formula). The data returned by FF_GEOREV_EX_DOMICILE_PCT is opposite to FF_GEOREV_COUNTRY_PCT ( which gives the percentage of sales generated from a specific country). In other words: FF_GEOREV_EX_DOMICILE_PCT = 100 - FF_GEOREV_COUNTRY_PCT. Although FF_GEOREV_EX_DOMICILE_PCT allows time series capability, it does not work. Treat this formula as a single point formula where you can specify a specific year of data.<br>Example Question: Do you have a code that tells me the percentage of a company's sales generated outside a specific country?<br>FactSet Description: </code>	<code>0.0</code>

• Loss: <code>FitMixinLoss</code>

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• fp16: True

All Hyperparameters

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

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 8
• per_device_eval_batch_size: 8
• 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: 3
• 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: True
• 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: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• fsdp_transformer_layer_cls_to_wrap: None
• 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
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• 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
• hub_revision: None
• 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
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 5.1.0
• Transformers: 4.56.1
• PyTorch: 2.8.0.dev20250319+cu128
• Accelerate: 1.10.1
• Datasets: 4.1.0
• Tokenizers: 0.22.0

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",
}

<!-- ## 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.* -->