/ Модель

Характеристики

Контекст
8 192 токена
Параметры
149 млн
Архитектура
modernbert · dense
Базовая модель
ModernBERT-base
Библиотека
sentence-transformers
Опубликована
14 фев 2025

reranker-modernbert-base-msmarco-bce

Классификация текстаТекст → label9 профессий

Англоязычный CrossEncoder на базе ModernBERT для оценки релевантности пар текстов и повторного ранжирования результатов поиска.

240скачиваний за 30 дней0лайковвеса обновлены14 фев 2025

Бенчмарки

Как модель справилась с задачами пользователей ailist и что о ней публикуют открытые лидерборды

Публичные оценки

Из открытых лидербордов. Сравнивайте результаты одного теста при одинаковых условиях — отсутствие оценки не значит, что модель хуже.

В подключённых источниках результатов для этой модели пока нет.

Что умеет модель

Описание составлено по карточке модели и метаданным Hugging Face

tomaarsen выпустил англоязычный CrossEncoder на базе answerdotai/ModernBERT-base, дообученный библиотекой sentence-transformers на парах запрос–ответ с бинарными метками релевантности. Модель содержит 149 млн параметров и обрабатывает последовательности длиной до 8192 токенов; для каждой пары текстов она выдаёт оценку, пригодную для ранжирования и оценки сходства.

Модель предназначена прежде всего для повторного ранжирования результатов поиска и семантического поиска, а также может использоваться для оценки сходства текстов, поиска перефразировок, классификации и кластеризации. В README приведены результаты оценки на NanoMSMARCO, NanoNFCorpus, NanoNQ и NanoBEIR. Для применения подходят сценарии, где нужно оценивать пару «запрос — документ» или сравнивать два текста.

Это текстовая encoder-модель, а не генератор ответов: она оценивает пары входных текстов. Обучающие данные в карточке описаны как набор из 19,99 млн пар query/answer с бинарными метками. Условия лицензии в предоставленных данных не указаны.

Сценарии применения

  • повторное ранжирование документов по запросу
  • оценка релевантности ответа поисковому запросу
  • семантический поиск по текстовым документам
  • оценка семантического сходства двух текстов
  • поиск перефразировок

Ограничения

Модель оценивает пары текстов и сама не генерирует ответы; основное обучение и приведённые оценки ориентированы на англоязычный поиск. В предоставленных данных лицензия не указана.

Для каких профессий подходит9

Разделы каталога профессий, в задачах которых используется эта модель

Все профессии

Также подходит для

Карточка модели

README от автора на Hugging Face, как он его опубликовал

Открыть на Hugging Face

CrossEncoder based on answerdotai/ModernBERT-base

This is a Cross Encoder model finetuned from answerdotai/ModernBERT-base using the sentence-transformers library. It computes scores for pairs of texts, which can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

  • Model Type: Cross Encoder
  • Base model: answerdotai/ModernBERT-base
  • Maximum Sequence Length: 8192 tokens
  • Number of Output Labels: 1 label

Model Sources

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("tomaarsen/reranker-modernbert-base-msmarco-bce")
# Get scores for pairs of texts
pairs = [
    ['what gb sp model is bright', 'From the list above it is easy to understand that Gegabyte is bigger than Megabyte. Or GB is bigger between MB and GB. Thanks.'],
    ['does immunotherapy work', 'The US and Drug Administration (FDA) this week convened a panel of outside experts to weigh in on the readiness of a first-of-its-kind cancer therapy. The treatment, which works by tweaking a patientâ\x80\x99s own cells, is a type of immunotherapy called CAR T-cell therapy and has been in clinical trials for several years. One drug maker is now seeking FDA approval to use the treatment in pediatric and young adult patients ages 3 to 25 with B-cell acute lymphoblastic leukemia (ALL) that has not responded to standard care.'],
    ['how long to wear oasis contacts', 'There is something wrong with my Xperia Z last week, and I did a resetting to make the phone become original, but I forgot backing up some important contacts, so I have to find the way for Sony Xperia Z contacts recovery, finanlly, I got this Android Data Recovery software to recover lost contacts from my phone.here is something wrong with my Xperia Z last week, and I did a resetting to make the phone become original, but I forgot backing up some important contacts, so I have to find the way for Sony Xperia Z contacts recovery, finanlly, I got this Android Data Recovery software to recover lost contacts from my phone.'],
    ['water baby definition', "Someone very comfortable in the water, Good swimmers, and never scared while in bodies of water. There's Jadine, back in the lake. She's such a water baby. #water #aqua #babies #water babies #water kids."],
    ['youngest suicide case', 'Samantha Kuberskki was found hanging by a belt at her home in Oregon after being sent to her room for arguing with her mother. A six-year-old girl who was sent to her room for punishment is feared to be one of the youngest people to have ever committed suicide in the U.S. Samantha Kuberskki was found hanging by a belt at her home in Oregon after being sent to her room for arguing with her mother. Her death was ruled as suicide by the coroner - sparking a bitter row with police who investigated her death and insist it was an accident.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'what gb sp model is bright',
    [
        'From the list above it is easy to understand that Gegabyte is bigger than Megabyte. Or GB is bigger between MB and GB. Thanks.',
        'The US and Drug Administration (FDA) this week convened a panel of outside experts to weigh in on the readiness of a first-of-its-kind cancer therapy. The treatment, which works by tweaking a patientâ\x80\x99s own cells, is a type of immunotherapy called CAR T-cell therapy and has been in clinical trials for several years. One drug maker is now seeking FDA approval to use the treatment in pediatric and young adult patients ages 3 to 25 with B-cell acute lymphoblastic leukemia (ALL) that has not responded to standard care.',
        'There is something wrong with my Xperia Z last week, and I did a resetting to make the phone become original, but I forgot backing up some important contacts, so I have to find the way for Sony Xperia Z contacts recovery, finanlly, I got this Android Data Recovery software to recover lost contacts from my phone.here is something wrong with my Xperia Z last week, and I did a resetting to make the phone become original, but I forgot backing up some important contacts, so I have to find the way for Sony Xperia Z contacts recovery, finanlly, I got this Android Data Recovery software to recover lost contacts from my phone.',
        "Someone very comfortable in the water, Good swimmers, and never scared while in bodies of water. There's Jadine, back in the lake. She's such a water baby. #water #aqua #babies #water babies #water kids.",
        'Samantha Kuberskki was found hanging by a belt at her home in Oregon after being sent to her room for arguing with her mother. A six-year-old girl who was sent to her room for punishment is feared to be one of the youngest people to have ever committed suicide in the U.S. Samantha Kuberskki was found hanging by a belt at her home in Oregon after being sent to her room for arguing with her mother. Her death was ruled as suicide by the coroner - sparking a bitter row with police who investigated her death and insist it was an accident.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Reranking

MetricNanoMSMARCONanoNFCorpusNanoNQ
map0.6519 (+0.1623)0.3432 (+0.0728)0.6951 (+0.2744)
mrr@100.6449 (+0.1674)0.5016 (+0.0017)0.7152 (+0.2885)
ndcg@100.7069 (+0.1665)0.3801 (+0.0550)0.7469 (+0.2462)

Cross Encoder Nano BEIR

MetricValue
map0.5634 (+0.1698)
mrr@100.6206 (+0.1525)
ndcg@100.6113 (+0.1559)

Training Details

Training Dataset

Unnamed Dataset

  • Size: 19,990,000 training samples
  • Columns: query, answer, and label
  • Approximate statistics based on the first 1000 samples:
    queryanswerlabel
    typestringstringint
    detailsmin: 10 charactersmean: 34.21 charactersmax: 197 charactersmin: 82 charactersmean: 350.38 charactersmax: 860 characters0: ~73.10%1: ~26.90%
  • Samples:
    queryanswerlabel
    who plays the trickster on flashThe Flash (2014 TV series) The Flash is a TV show based on the fictional character Flash, a costumed superhero crime-fighter who appears in comic books published by DC Comics.0
    what type of business is plastics engineering companyPlastics Engineering Company is a leading North American manufacturer of phenolic resins and thermoset molding materials, selling products under its trademark Plenco. If you have a phenolic resin or thermoset molding material project, chances are, the Plenco team can make it work. We've been doing it for over 80 years. Come and benefit from the Plenco difference. Plastics Engineering Company, a family owned and managed business founded in 1934, established as its corporate mission a sincere desire to respond efficiently to the needs of our customers through development, manufacture, and servicing of useful, high-value products.1
    what is allianz global assistancePlease choose 'Allianz Direct Customers' for Car, Home, Pet, Boat and Horse & Rider Insurance. Allianz Direct Customers Allianz Direct Customers Car, Home, Pet, Boat and Horse & Rider Insurance. Phone. In the Republic of Ireland: 01 448 48 48. Outside Republic of Ireland: 00 353 1 448 48 48. Opening Hours: Monday to Friday 8am - 6pm and Saturday 9am - 1pm.0
  • Loss: BinaryCrossEntropyLoss with these parameters:
    {
        "activation_fct": "Identity",
        "pos_weight": 4
    }
    

Evaluation Dataset

Unnamed Dataset

  • Size: 10,000 evaluation samples
  • Columns: query, answer, and label
  • Approximate statistics based on the first 1000 samples:
    queryanswerlabel
    typestringstringint
    detailsmin: 11 charactersmean: 33.77 charactersmax: 215 charactersmin: 73 charactersmean: 351.17 charactersmax: 935 characters0: ~75.80%1: ~24.20%
  • Samples:
    queryanswerlabel
    what gb sp model is brightFrom the list above it is easy to understand that Gegabyte is bigger than Megabyte. Or GB is bigger between MB and GB. Thanks.0
    does immunotherapy workThe US and Drug Administration (FDA) this week convened a panel of outside experts to weigh in on the readiness of a first-of-its-kind cancer therapy. The treatment, which works by tweaking a patient’s own cells, is a type of immunotherapy called CAR T-cell therapy and has been in clinical trials for several years. One drug maker is now seeking FDA approval to use the treatment in pediatric and young adult patients ages 3 to 25 with B-cell acute lymphoblastic leukemia (ALL) that has not responded to standard care.0
    how long to wear oasis contactsThere is something wrong with my Xperia Z last week, and I did a resetting to make the phone become original, but I forgot backing up some important contacts, so I have to find the way for Sony Xperia Z contacts recovery, finanlly, I got this Android Data Recovery software to recover lost contacts from my phone.here is something wrong with my Xperia Z last week, and I did a resetting to make the phone become original, but I forgot backing up some important contacts, so I have to find the way for Sony Xperia Z contacts recovery, finanlly, I got this Android Data Recovery software to recover lost contacts from my phone.0
  • Loss: BinaryCrossEntropyLoss with these parameters:
    {
        "activation_fct": "Identity",
        "pos_weight": 4
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 128
  • per_device_eval_batch_size: 128
  • learning_rate: 8e-05
  • num_train_epochs: 1
  • warmup_ratio: 0.1
  • seed: 12
  • bf16: True
  • dataloader_num_workers: 4
  • load_best_model_at_end: True

All Hyperparameters

  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 128
  • per_device_eval_batch_size: 128
  • 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: 8e-05
  • weight_decay: 0.0
  • adam_beta1: 0.9
  • adam_beta2: 0.999
  • adam_epsilon: 1e-08
  • max_grad_norm: 1.0
  • num_train_epochs: 1
  • max_steps: -1
  • lr_scheduler_type: linear
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.1
  • 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: 12
  • data_seed: None
  • jit_mode_eval: False
  • use_ipex: False
  • bf16: True
  • 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: 4
  • dataloader_prefetch_factor: None
  • past_index: -1
  • disable_tqdm: False
  • remove_unused_columns: True
  • label_names: None
  • load_best_model_at_end: True
  • 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}
  • 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
  • dispatch_batches: None
  • split_batches: 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: proportional

Training Logs

EpochStepTraining LossValidation LossNanoMSMARCO_ndcg@10NanoNFCorpus_ndcg@10NanoNQ_ndcg@10NanoBEIR_mean_ndcg@10
-1-1--0.0245 (-0.5159)0.2709 (-0.0541)0.0858 (-0.4148)0.1271 (-0.3283)
0.000011.1359-----
0.006410000.9962-----
0.012820000.3958-----
0.019230000.3292-----
0.025640000.3023-----
0.032050000.28390.24950.6231 (+0.0827)0.3748 (+0.0498)0.7295 (+0.2288)0.5758 (+0.1204)
0.038460000.2812-----
0.044870000.2755-----
0.051280000.2672-----
0.057690000.2624-----
0.0640100000.26270.23680.6976 (+0.1572)0.4094 (+0.0844)0.7139 (+0.2133)0.6070 (+0.1516)
0.0704110000.2586-----
0.0768120000.253-----
0.0832130000.25-----
0.0896140000.2545-----
0.0960150000.25160.22970.6671 (+0.1267)0.3685 (+0.0434)0.7301 (+0.2295)0.5886 (+0.1332)
0.1025160000.241-----
0.1089170000.2459-----
0.1153180000.2371-----
0.1217190000.2374-----
0.1281200000.2340.22260.6377 (+0.0973)0.3988 (+0.0737)0.7002 (+0.1996)0.5789 (+0.1235)
0.1345210000.2293-----
0.1409220000.2222-----
0.1473230000.2231-----
0.1537240000.2212-----
0.1601250000.21650.22660.7114 (+0.1710)0.3775 (+0.0524)0.7314 (+0.2308)0.6068 (+0.1514)
0.1665260000.2119-----
0.1729270000.2086-----
0.1793280000.204-----
0.1857290000.204-----
0.1921300000.19590.19130.6630 (+0.1225)0.3962 (+0.0712)0.7027 (+0.2020)0.5873 (+0.1319)
0.1985310000.195-----
0.2049320000.1899-----
0.2113330000.1887-----
0.2177340000.1865-----
0.2241350000.18780.17650.6709 (+0.1304)0.3858 (+0.0607)0.7060 (+0.2053)0.5875 (+0.1322)
0.2305360000.1822-----
0.2369370000.1795-----
0.2433380000.1802-----
0.2497390000.1762-----
0.2561400000.16940.17390.6902 (+0.1498)0.3771 (+0.0521)0.7198 (+0.2192)0.5957 (+0.1403)
0.2625410000.1718-----
0.2689420000.1706-----
0.2753430000.1659-----
0.2817440000.1593-----
0.2881450000.16080.15320.7132 (+0.1728)0.3606 (+0.0356)0.7393 (+0.2386)0.6044 (+0.1490)
0.2945460000.1589-----
0.3010470000.1563-----
0.3074480000.1553-----
0.3138490000.155-----
0.3202500000.15010.13730.7168 (+0.1764)0.3830 (+0.0579)0.6954 (+0.1948)0.5984 (+0.1430)
0.3266510000.1508-----
0.3330520000.1497-----
0.3394530000.1478-----
0.3458540000.1445-----
0.3522550000.14680.14030.6828 (+0.1424)0.3780 (+0.0530)0.7147 (+0.2141)0.5919 (+0.1365)
0.3586560000.1422-----
0.3650570000.1369-----
0.3714580000.1364-----
0.3778590000.1328-----
0.3842600000.13510.14480.6881 (+0.1477)0.3430 (+0.0179)0.7267 (+0.2260)0.5859 (+0.1306)
0.3906610000.1312-----
0.3970620000.1308-----
0.4034630000.1289-----
0.4098640000.1273-----
0.4162650000.12570.12900.7288 (+0.1883)0.3830 (+0.0580)0.7180 (+0.2173)0.6099 (+0.1545)
0.4226660000.1246-----
0.4290670000.1275-----
0.4354680000.1246-----
0.4418690000.1214-----
0.4482700000.1150.11840.6911 (+0.1506)0.3903 (+0.0652)0.7189 (+0.2182)0.6001 (+0.1447)
0.4546710000.113-----
0.4610720000.1156-----
0.4674730000.1142-----
0.4738740000.1133-----
0.4802750000.11320.11940.7069 (+0.1665)0.3801 (+0.0550)0.7469 (+0.2462)0.6113 (+0.1559)
0.4866760000.1085-----
0.4930770000.1095-----
0.4994780000.1105-----
0.5059790000.1068-----
0.5123800000.10390.10850.7017 (+0.1612)0.3565 (+0.0315)0.7199 (+0.2192)0.5927 (+0.1373)
0.5187810000.1059-----
0.5251820000.1001-----
0.5315830000.1019-----
0.5379840000.1021-----
0.5443850000.09820.09620.6842 (+0.1438)0.3516 (+0.0266)0.7431 (+0.2425)0.5930 (+0.1376)
0.5507860000.0967-----
0.5571870000.0962-----
0.5635880000.098-----
0.5699890000.0973-----
0.5763900000.09570.08630.6729 (+0.1325)0.3852 (+0.0601)0.7147 (+0.2141)0.5909 (+0.1356)
0.5827910000.0925-----
0.5891920000.0948-----
0.5955930000.0887-----
0.6019940000.0918-----
0.6083950000.09260.08460.6857 (+0.1453)0.3503 (+0.0253)0.7321 (+0.2315)0.5894 (+0.1340)
0.6147960000.0881-----
0.6211970000.0871-----
0.6275980000.0867-----
0.6339990000.0854-----
0.64031000000.08330.07900.6665 (+0.1261)0.3415 (+0.0165)0.6905 (+0.1898)0.5662 (+0.1108)
0.64671010000.0837-----
0.65311020000.0834-----
0.65951030000.0798-----
0.66591040000.0825-----
0.67231050000.08030.07500.6897 (+0.1493)0.3415 (+0.0165)0.7096 (+0.2090)0.5803 (+0.1249)
0.67871060000.076-----
0.68511070000.0782-----
0.69151080000.0786-----
0.69791090000.075-----
0.70441100000.07470.06900.6665 (+0.1261)0.3384 (+0.0134)0.7209 (+0.2202)0.5753 (+0.1199)
0.71081110000.0728-----
0.71721120000.0708-----
0.72361130000.0714-----
0.73001140000.0725-----
0.73641150000.07080.06590.6753 (+0.1348)0.3423 (+0.0172)0.7093 (+0.2087)0.5756 (+0.1202)
0.74281160000.0684-----
0.74921170000.0709-----
0.75561180000.0661-----
0.76201190000.0685-----
0.76841200000.06550.06130.6774 (+0.1369)0.3295 (+0.0044)0.7244 (+0.2238)0.5771 (+0.1217)
0.77481210000.0643-----
0.78121220000.066-----
0.78761230000.0625-----
0.79401240000.0653-----
0.80041250000.06190.05640.6797 (+0.1393)0.3598 (+0.0348)0.7193 (+0.2187)0.5863 (+0.1309)
0.80681260000.0616-----
0.81321270000.0607-----
0.81961280000.0584-----
0.82601290000.0609-----
0.83241300000.05680.05020.6855 (+0.1450)0.3394 (+0.0143)0.7297 (+0.2291)0.5849 (+0.1295)
0.83881310000.0577-----
0.84521320000.056-----
0.85161330000.0556-----
0.85801340000.0553-----
0.86441350000.05460.04710.6903 (+0.1499)0.3404 (+0.0153)0.7419 (+0.2413)0.5909 (+0.1355)
0.87081360000.0525-----
0.87721370000.0512-----
0.88361380000.0528-----
0.89001390000.0523-----
0.89641400000.05440.04420.6915 (+0.1511)0.3507 (+0.0257)0.7258 (+0.2251)0.5893 (+0.1340)
0.90291410000.0497-----
0.90931420000.0508-----
0.91571430000.0485-----
0.92211440000.0492-----
0.92851450000.04720.04420.6614 (+0.1210)0.3394 (+0.0144)0.7361 (+0.2355)0.5790 (+0.1236)
0.93491460000.0469-----
0.94131470000.0459-----
0.94771480000.0471-----
0.95411490000.0454-----
0.96051500000.04440.04290.6587 (+0.1183)0.3311 (+0.0060)0.7298 (+0.2291)0.5732 (+0.1178)
0.96691510000.0451-----
0.97331520000.0429-----
0.97971530000.0448-----
0.98611540000.0441-----
0.99251550000.04430.04180.6653 (+0.1249)0.3335 (+0.0084)0.7391 (+0.2385)0.5793 (+0.1239)
0.99891560000.0409-----
-1-1--0.7069 (+0.1665)0.3801 (+0.0550)0.7469 (+0.2462)0.6113 (+0.1559)
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.11.10
  • Sentence Transformers: 3.5.0.dev0
  • Transformers: 4.49.0.dev0
  • PyTorch: 2.6.0.dev20241112+cu121
  • Accelerate: 1.2.0
  • Datasets: 3.2.0
  • Tokenizers: 0.21.0

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",
}

Как запустить

На своей GPU в ailist. Цена — по текущим предложениям провайдеров

GPU в ailist

Универсальный рантайм · sentence-transformers · text-classification

Можно запустить

Подбираем GPU…

Класс архитектуры
ModernBertForSequenceClassification
Ревизия весов
1d84d75

0 моделей