KR20240089615A - 사전-트레이닝된 언어 모델의 단일 트랜스포머 계층으로부터의 다중-헤드 네트워크의 미세-튜닝 - Google Patents

사전-트레이닝된 언어 모델의 단일 트랜스포머 계층으로부터의 다중-헤드 네트워크의 미세-튜닝 Download PDF

Info

Publication number
KR20240089615A
KR20240089615A KR1020247015690A KR20247015690A KR20240089615A KR 20240089615 A KR20240089615 A KR 20240089615A KR 1020247015690 A KR1020247015690 A KR 1020247015690A KR 20247015690 A KR20247015690 A KR 20247015690A KR 20240089615 A KR20240089615 A KR 20240089615A
Authority
KR
South Korea
Prior art keywords
layers
bot
machine
learning model
fine
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
KR1020247015690A
Other languages
English (en)
Korean (ko)
Inventor
탄 티엔 부
투옌 꽝 팜
오미드 모하마드 네자미
마크 에드워드 존슨
탄 롱 동
콩 주이 부 호앙
Original Assignee
오라클 인터내셔날 코포레이션
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 오라클 인터내셔날 코포레이션 filed Critical 오라클 인터내셔날 코포레이션
Publication of KR20240089615A publication Critical patent/KR20240089615A/ko
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
    • G10L15/063—Training
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0475—Generative networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/096—Transfer learning
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
    • G10L15/063—Training
    • G10L2015/0635—Training updating or merging of old and new templates; Mean values; Weighting
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/223—Execution procedure of a spoken command

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Acoustics & Sound (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Machine Translation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
KR1020247015690A 2021-10-12 2022-08-17 사전-트레이닝된 언어 모델의 단일 트랜스포머 계층으로부터의 다중-헤드 네트워크의 미세-튜닝 Pending KR20240089615A (ko)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
US202163254740P 2021-10-12 2021-10-12
US63/254,740 2021-10-12
US17/735,651 2022-05-03
US17/735,651 US12512091B2 (en) 2021-10-12 2022-05-03 Fine-tuning multi-head network from a single transformer layer of pre-trained language model
PCT/US2022/040530 WO2023064033A1 (en) 2021-10-12 2022-08-17 Fine-tuning multi-head network from a single transformer layer of pre-trained language model

Publications (1)

Publication Number Publication Date
KR20240089615A true KR20240089615A (ko) 2024-06-20

Family

ID=85798249

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020247015690A Pending KR20240089615A (ko) 2021-10-12 2022-08-17 사전-트레이닝된 언어 모델의 단일 트랜스포머 계층으로부터의 다중-헤드 네트워크의 미세-튜닝

Country Status (6)

Country Link
US (2) US12512091B2 (https=)
JP (1) JP2024539003A (https=)
KR (1) KR20240089615A (https=)
CN (1) CN118140230A (https=)
GB (1) GB2631139A (https=)
WO (1) WO2023064033A1 (https=)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12548552B2 (en) * 2021-11-19 2026-02-10 International Business Machines Corporation Dynamic language selection of an AI voice assistance system
US11947935B2 (en) * 2021-11-24 2024-04-02 Microsoft Technology Licensing, Llc. Custom models for source code generation via prefix-tuning
US20240061835A1 (en) * 2022-08-22 2024-02-22 Oracle International Corporation System and method of selective fine-tuning for custom training of a natural language to logical form model
US20240169165A1 (en) * 2022-11-17 2024-05-23 Samsung Electronics Co., Ltd. Automatically Generating Annotated Ground-Truth Corpus for Training NLU Model
US12562163B2 (en) * 2023-05-12 2026-02-24 Servicenow, Inc. Bidirectional assistant for development platforms
CN116774140A (zh) * 2023-06-26 2023-09-19 南京邮电大学 基于残差注意力网络的无网格信号源doa估计方法
US20250005282A1 (en) * 2023-06-29 2025-01-02 Amazon Technologies, Inc. Domain entity extraction for performing text analysis tasks
CN118446218B (zh) * 2024-05-16 2024-11-01 西南交通大学 一种对抗式阅读理解嵌套命名实体识别方法
CA3253531A1 (en) * 2024-06-14 2026-01-19 The Toronto-Dominion Bank Context retrieval for in-context learning model
WO2026000314A1 (en) * 2024-06-27 2026-01-02 Beijing Youzhuju Network Technology Co., Ltd. Model-based task processing
JP7658644B1 (ja) * 2024-10-21 2025-04-08 スパーブエーアイ カンパニー リミテッド 事前学習されたベースモデルに基づいたカスタムモデルを学習する方法及びそれを用いた学習装置｛ｍｅｔｈｏｄ ｆｏｒ ｔｒａｉｎｉｎｇ ｃｕｓｔｏｍ ｍｏｄｅｌ ｂａｓｅｄ ｏｎ ｐｒｅ－ｔｒａｉｎｅｄ ｂａｓｅ ｍｏｄｅｌ ａｎｄ ｌｅａｒｎｉｎｇ ｄｅｖｉｃｅ ｕｓｉｎｇ ｔｈｅ ｓａｍｅ｝
CN119418321B (zh) * 2024-10-30 2025-09-30 上海哔哩哔哩科技有限公司 模型训练方法、用于检测和识别文本的方法及相关装置
CN119418319B (zh) * 2024-10-30 2025-09-30 上海哔哩哔哩科技有限公司 模型训练方法、文本检测方法、装置、介质和程序产品
CN119418320B (zh) * 2024-10-30 2025-09-30 上海哔哩哔哩科技有限公司 一种模型训练方法、装置、介质和程序产品
CN119915374B (zh) * 2025-04-03 2025-11-14 浙江潮汐力科技有限公司 故障监测方法、装置、设备、存储介质和程序产品

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11138392B2 (en) * 2018-07-26 2021-10-05 Google Llc Machine translation using neural network models
US20200042864A1 (en) 2018-08-02 2020-02-06 Veritone, Inc. Neural network orchestration
US11556778B2 (en) 2018-12-07 2023-01-17 Microsoft Technology Licensing, Llc Automated generation of machine learning models
US20210279596A1 (en) 2020-03-06 2021-09-09 Hitachi, Ltd. System for predictive maintenance using trace norm generative adversarial networks
US20220094713A1 (en) * 2020-09-21 2022-03-24 Sophos Limited Malicious message detection
US12141701B2 (en) * 2021-01-21 2024-11-12 International Business Machines Corporation Channel scaling: a scale-and-select approach for selective transfer learning
US11875898B2 (en) * 2021-05-26 2024-01-16 Merative Us L.P. Automatic condition diagnosis using an attention-guided framework
US20230106669A1 (en) * 2021-09-27 2023-04-06 X Development Llc Binding affinity prediction using neural networks

Also Published As

Publication number Publication date
US20260080864A1 (en) 2026-03-19
GB2631139A (en) 2024-12-25
JP2024539003A (ja) 2024-10-28
US12512091B2 (en) 2025-12-30
CN118140230A (zh) 2024-06-04
US20230115321A1 (en) 2023-04-13
GB202403625D0 (en) 2024-04-24
WO2023064033A1 (en) 2023-04-20

Similar Documents

Publication Publication Date Title
US12299402B2 (en) Techniques for out-of-domain (OOD) detection
US12361219B2 (en) Context tag integration with named entity recognition models
JP7561836B2 (ja) 自然言語処理のためのストップワードデータ拡張
US12099816B2 (en) Multi-factor modelling for natural language processing
US12512091B2 (en) Fine-tuning multi-head network from a single transformer layer of pre-trained language model
JP2025530712A (ja) 自然言語インターフェースとして訓練されている機械学習モデルの信頼スコアの較正
US12412563B2 (en) Path dropout for natural language processing
KR102821062B1 (ko) 사전-트레이닝된 언어 모델들에 대한 긴 텍스트를 핸들링하기 위한 시스템 및 기술들
US12572852B2 (en) Lexical dropout for natural language processing
KR20250029146A (ko) 개체-인식 데이터 증강을 위한 기술들
KR20240096829A (ko) 해시 임베딩들을 사용하는 언어 검출을 위한 와이드 및 딥 네트워크
US20230136965A1 (en) Prohibiting inconsistent named entity recognition tag sequences
US20260065171A1 (en) Adaptive training data augmentation to facilitate training named entity recognition models

Legal Events

Date Code Title Description
PA0105 International application

Patent event date: 20240510

Patent event code: PA01051R01D

Comment text: International Patent Application

PG1501 Laying open of application