Bloomz - 560m Open - Source Multilingual Model - The Top Choice for Multi - Scenario Natural Language Processing Tasks

Bloomz 560m

Developed by bigscience

A medium-sized multilingual model from the BLOOMZ series, suitable for various natural language processing tasks.

Supports Multiple LanguagesOpen Source License:Openrail #Multilingual text generation #Cross-lingual sentiment analysis #Low-resource language support

Downloads 593.72k

Release Time : 10/8/2022

Model Overview

BLOOMZ-560M is a multilingual text generation model fine-tuned from the BLOOM large language model, supporting multiple languages and programming language comprehension and generation.

Model Features

Multilingual Support

Supports over 46 natural languages and multiple programming languages, suitable for international applications.

Strong Text Generation Capability

Performs well on various tasks such as fable creation, question answering, sentiment analysis, and other generative tasks.

Cross-lingual Transfer Ability

Capable of cross-lingual understanding and generation, enabling interoperability between multiple languages.

Open-source and Extensible

The model and training datasets are publicly available, facilitating further optimization and application development by researchers and developers.

Model Capabilities

Text generation

Multilingual translation

Sentiment analysis

Question answering system

Programming code generation

Coreference resolution

Natural language inference

Search keyword recommendation

Use Cases

Content Creation

Fable Generation

Generate fables with specific morals in Spanish or Hindi.

Outputs structurally complete, linguistically fluent fables that align with cultural contexts.

Fairy Tale Writing

Generate globally influential fairy tales based on specified plots.

Produces educational and widely recognized literary works.

Language Understanding

Sentiment Analysis

Determine the stance (positive/neutral/negative) of mixed Chinese-English reviews.

Accurately identifies emotional tendencies in the text.

Cross-lingual Question Answering

Explain the backpropagation principle in neural networks using Telugu.

Provides clear and accurate technical term explanations.

Information Retrieval

Keyword Suggestion

Generate related search terms for the Vietnamese query 'artificial neural network'.

Outputs high-quality, semantically relevant keyword lists.

French Term Expansion

Propose multiple French keywords around the theme of 'artificial neural network'.

Helps users discover terms and concepts in related fields.

🚀 Bloomz-560m Model Introduction

This document provides detailed information about the bloomz-560m model, including its datasets, supported languages, programming languages, and performance metrics.

📚 Documentation

Datasets

The model is trained on the following dataset:

bigscience/xP3

Supported Languages

The model supports a wide range of languages, including:

ak, ar, as, bm, bn, ca, code, en, es, eu, fon, fr, gu, hi, id, ig, ki, kn, lg, ln, ml, mr, ne, nso, ny, or, pa, pt, rn, rw, sn, st, sw, ta, te, tn, ts, tum, tw, ur, vi, wo, xh, yo, zh, zu

Programming Languages

The model is related to the following programming languages:

C, C++, C#, Go, Java, JavaScript, Lua, PHP, Python, Ruby, Rust, Scala, TypeScript

Pipeline Tag

The pipeline tag for this model is text-generation.

Widget Examples

Here are some widget examples demonstrating different tasks:

Example Title	Text
"zh-en sentiment"	"A legendary beginning, an immortal myth. This is not just a movie, but a label for entering a new era, forever inscribed in the annals of history. Would you rate the previous review as positive, neutral or negative?"
"zh-zh sentiment"	"一个传奇的开端，一个不灭的神话，这不仅仅是一部电影，而是作为一个走进新时代的标签，永远彪炳史册。你认为这句话的立场是赞扬、中立还是批评？"
"vi-en query"	"Suggest at least five related search terms to "Mạng neural nhân tạo"."
"fr-fr query"	"Proposez au moins cinq mots clés concernant «Réseau de neurones artificiels»."
"te-en qa"	"Explain in a sentence in Telugu what is backpropagation in neural networks."
"en-en qa"	"Why is the sky blue?"
"es-en fable"	"Write a fairy tale about a troll saving a princess from a dangerous dragon. The fairy tale is a masterpiece that has achieved praise worldwide and its moral is "Heroes Come in All Shapes and Sizes". Story (in Spanish):"
"hi-en fable"	"Write a fable about wood elves living in a forest that is suddenly invaded by ogres. The fable is a masterpiece that has achieved praise worldwide and its moral is "Violence is the last refuge of the incompetent". Fable (in Hindi):"

Model Results

The following table presents the performance metrics of the bloomz-560m model on various tasks and datasets:

Task Type	Dataset Name	Accuracy / Pass@N	Value
Coreference resolution	Winogrande XL (xl)	Accuracy	52.41
Coreference resolution	XWinograd (en)	Accuracy	51.01
Coreference resolution	XWinograd (fr)	Accuracy	51.81
Coreference resolution	XWinograd (jp)	Accuracy	52.03
Coreference resolution	XWinograd (pt)	Accuracy	53.99
Coreference resolution	XWinograd (ru)	Accuracy	53.97
Coreference resolution	XWinograd (zh)	Accuracy	54.76
Natural language inference	ANLI (r1)	Accuracy	33.4
Natural language inference	ANLI (r2)	Accuracy	33.4
Natural language inference	ANLI (r3)	Accuracy	33.5
Natural language inference	SuperGLUE (cb)	Accuracy	53.57
Natural language inference	SuperGLUE (rte)	Accuracy	67.15
Natural language inference	XNLI (ar)	Accuracy	44.46
Natural language inference	XNLI (bg)	Accuracy	39.76
Natural language inference	XNLI (de)	Accuracy	39.36
Natural language inference	XNLI (el)	Accuracy	40.96
Natural language inference	XNLI (en)	Accuracy	46.43
Natural language inference	XNLI (es)	Accuracy	44.98
Natural language inference	XNLI (fr)	Accuracy	45.54
Natural language inference	XNLI (hi)	Accuracy	41.81
Natural language inference	XNLI (ru)	Accuracy	39.64
Natural language inference	XNLI (sw)	Accuracy	38.35
Natural language inference	XNLI (th)	Accuracy	35.5
Natural language inference	XNLI (tr)	Accuracy	37.31
Natural language inference	XNLI (ur)	Accuracy	38.96
Natural language inference	XNLI (vi)	Accuracy	44.74
Natural language inference	XNLI (zh)	Accuracy	44.66
Program synthesis	HumanEval	Pass@1	2.18
Program synthesis	HumanEval	Pass@10	4.11
Program synthesis	HumanEval	Pass@100	9.00
Sentence completion	StoryCloze (2016)	Accuracy	60.29
Sentence completion	SuperGLUE (copa)	Accuracy	52.0
Sentence completion	XCOPA (et)	Accuracy	53.0
Sentence completion	XCOPA (ht)	Accuracy	49.0
Sentence completion	XCOPA (id)	Accuracy	57.0
Sentence completion	XCOPA (it)	Accuracy	52.0
Sentence completion	XCOPA (qu)	Accuracy	55.0
Sentence completion	XCOPA (sw)	Accuracy	56.0
Sentence completion	XCOPA (ta)	Accuracy	58.0
Sentence completion	XCOPA (th)	Accuracy	58.0
Sentence completion	XCOPA (tr)	Accuracy	61.0
Sentence completion	XCOPA (vi)	Accuracy	61.0
Sentence completion	XCOPA (zh)	Accuracy	61.0
Sentence completion	XStoryCloze (ar)	Accuracy	54.4
Sentence completion	XStoryCloze (es)	Accuracy	56.45
Sentence completion	XStoryCloze (eu)	Accuracy	[Value not provided]

📄 License

The model is released under the bigscience-bloom-rail-1.0 license.

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