mbert-base-parsinlu-entailment Open-source Model - Supports Logical Judgment of Persian Multilingual Textual Entailment

Mbert Base Parsinlu Entailment

Developed by persiannlp

This is a model designed to handle text entailment problems, supporting logical relationship judgment in Persian and multilingual contexts.

Text Classification Other#Persian Text Entailment #Multilingual Logical Reasoning #Contradiction Detection

Downloads 39

Release Time : 3/2/2022

Model Overview

This model is specifically used to determine the logical relationship (entailment, contradiction, or neutrality) between two pieces of text. It is built based on the ParsBERT architecture and is suitable for Persian text analysis.

Model Features

Multilingual Support

Based on the mBERT architecture, it supports text entailment analysis in Persian and multiple other languages.

Logical Relationship Judgment

Capable of accurately identifying three types of logical relationships between texts: entailment, contradiction, and neutrality.

Pretrained Model

Based on the ParsBERT pretrained model, it performs excellently on Persian language tasks.

Model Capabilities

Text Logical Relationship Analysis

Persian Text Processing

Multilingual Text Understanding

Use Cases

Natural Language Processing

Text Logical Verification

Verify whether there is a logical entailment relationship between two pieces of text.

Can accurately determine the logical relationship between texts.

Question Answering Systems

Used in question-answering systems to verify the logical consistency between answers and questions.

Improves the accuracy of question-answering systems.

Content Moderation

Contradiction Content Detection

Detect whether there are self-contradictory statements in user-submitted content.

Helps identify false or misleading information.

🚀 Textual Entailment (Model for Textual Entailment Problems)

This is a model designed to solve textual entailment problems, offering a practical solution for related tasks.

🚀 Quick Start

This is a model for textual entailment problems. Here is an example of how you can run this model:

import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer
import numpy as np

labels = ["entails", "contradicts", "neutral"]
model_name_or_path = "persiannlp/mbert-base-parsinlu-entailment"
model = AutoModelForSequenceClassification.from_pretrained(model_name_or_path)
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path,)


def model_predict(text_a, text_b):
    features = tokenizer( [(text_a, text_b)], padding="max_length", truncation=True, return_tensors='pt')
    output = model(**features)
    logits = output[0]
    probs = torch.nn.functional.softmax(logits, dim=1).tolist()
    idx = np.argmax(np.array(probs))
    print(labels[idx], probs)


model_predict(
    "این مسابقات بین آوریل و دسامبر در هیپودروم ولیفندی در نزدیکی باکرکی ، ۱۵ کیلومتری (۹ مایل) غرب استانبول برگزار می شود.",
    "در ولیفندی هیپودروم، مسابقاتی از آوریل تا دسامبر وجود دارد."
)

model_predict(
"آیا کودکانی وجود دارند که نیاز به سرگرمی دارند؟",
    "هیچ کودکی هرگز نمی خواهد سرگرم شود.",
)

model_predict(
    "ما به سفرهایی رفته ایم که در نهرهایی شنا کرده ایم",
    "علاوه بر استحمام در نهرها ، ما به اسپا ها و سونا ها نیز رفته ایم."
)

📚 Documentation

For more details, visit this page: https://github.com/persiannlp/parsinlu/

📄 License

This project is licensed under the CC BY-NC-SA 4.0 license.

Property	Details
Language	Farsi, Multilingual
Thumbnail	https://upload.wikimedia.org/wikipedia/commons/a/a2/Farsi.svg
Tags	entailment, parsbert, persian, farsi
License	cc-by-nc-sa-4.0
Datasets	parsinlu
Metrics	accuracy

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご