Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · cointegrated

rubert-base-cased-nli-threeway

rubert-base-cased-nli-threeway is a Russian BERT NLI model from cointegrated with about 178M parameters. According to the model card, it predicts entailment, contradiction, or neutral text relationships.

Publisher
cointegrated
Task
zero-shot-classification
Model type
bert
License
Unknown
Library
transformers
Publication status
Accepted · not indexed

Model overview

rubert-base-cased-nli-threeway is published by cointegrated as a zero-shot-classification model. The captured configuration identifies BertForSequenceClassification with model type bert, and Safetensors metadata reports 177,856,259 parameters. According to the model card, it fine-tunes DeepPavlov/rubert-base-cased on Russian-translated NLI data to predict entailment, contradiction, or neutral.

Recorded capabilities

Three-way NLI

According to the model card, the model predicts entailment, contradiction, or neutral for pairs of short texts.

RuBERT base

According to the model card, it is DeepPavlov/rubert-base-cased fine-tuned for natural language inference.

Zero-shot classification path

According to the model card, the entailment label can score arbitrary label texts for zero-shot short text classification, with a Russian sentiment example and Hugging Face pipelines support.

Translated Russian NLI training

According to the model card, it was trained on NLI datasets automatically translated from English to Russian.

Use cases in the source record

  • Russian natural language inference on text pairs using the documented Transformers sequence-classification workflow.
  • Zero-shot short text classification by labels only, illustrated in the card with Russian sentiment examples and Hugging Face pipelines.

Limitations and unknowns

  • No license value was extracted from this record.
  • No context-window value was extracted from this record.
  • The card's evaluation table and extra-dataset list are publisher claims and were not independently verified.
  • Provider state is historical snapshot data and should be refreshed before being presented as current.

Source and provenance

Source: cointegrated/rubert-base-cased-nli-threeway

Captured: Unknown. Processed: 2026-09-07T19:34:42.777397+00:00.

RuBERT for NLI (natural language inference) This is the DeepPavlov/rubert-base-cased fine-tuned to predict the logical relationship between two short texts: entailment, contradiction, or neutral. Usage How to run the model for NLI: # !pip install transformers sentencepiece --quiet import torch from transformers import AutoTokenizer, AutoModelForSequenceClassification model_checkpoint = 'cointegrated/rubert-base-cased-nli-threeway' tokenizer = AutoTokenizer.from_pretrained(model_checkpoint) model = AutoModelForSequenceClassification.from_pretrained(model_checkpoint) if torch.cuda.is_available(): model.cuda() text1 = 'Сократ - человек…

F001F002F003F004F005F006F008F009F010F011F012F013F014F015F016