paper-with-me

홈 › Papers

Universal Language Model Fine-tuning for Text Classification

2018-01-18 · ACL 2018 7 · Jeremy Howard, Sebastian Ruder

Inductive transfer learning has greatly impacted computer vision, but existing approaches in NLP still require task-specific modifications and training from scratch. We propose Universal Language Model Fine-tuning (ULMFiT), an effective transfer learning method that can be applied to any task in NLP, and introduce techniques that are key for fine-tuning a language model. Our method significantly outperforms the state-of-the-art on six text classification tasks, reducing the error by 18-24% on the majority of datasets. Furthermore, with only 100 labeled examples, it matches the performance of training from scratch on 100x more data. We open-source our pretrained models and code.

📄 PDF Abstract BibTeX arXiv:1801.06146

Code (66)

fastai/fastai 공식 구현 pytorch
AbhimanyuAryan/IMDB-NLP
AnttiKarlsson/finnish_ulmfit pytorch
Deepayan137/Adapting-OCR pytorch
Julian1070/Deep-Learning pytorch
LamLauChiu/Tensorflow_Learning tf
MJahangeerQureshi/Text-Classification
Mees-Molenaar/protein_location pytorch
PrideLee/sentiment-analysis pytorch
RajasSU/Twitter-Sentiment-Analysis-using-ULMFiT
SkullFang/ULMFIT_NLP_Classification
Socialbird-AILab/BERT-Classification-Tutorial tf
TheShadow29/subreddit-classification-dataset tf
ahmadelsallab/READ tf
akari0216/Paddle-awdlstm paddle
alantancr/Hotel-Recommender tf
alexandra-chron/ntua-slp-wassa-iest2018 pytorch
alexandra-chron/wassa-2018 pytorch
algoashutosh/sentiment-analysis-project pytorch
amagooda/SummaRuNNer_coattention pytorch
anthonyckleung/Transfer-Learning-in-Sentiment-Tweets pytorch
anubhavmaity/Ag-News-Category-Classifier
apmoore1/language-model pytorch
benjaminvdb/110kDBRD
benjaminvdb/DBRD
bhuvanakundumani/sexist_rmks_classifier
cahya-wirawan/indonesian-language-models tf
castortroynz/desafio_atuacao19
comicencyclo/TransferLearning_DiscriminativeFineTuning
cstorm125/thai2fit pytorch
dhruvsawhney/CS152_FinalProject
dpalominop/ULMFit
floleuerer/fastai_ulmfit
floraxinru/HotelReviews_SentimentAnalysis
floraxinru/NLP_HotelReviews
haianhle/ULMFiT-Sentiment
heye0507/individualNLPClassifier pytorch
hpanwar08/sentence-classification-pytorch pytorch
jackbandy/deep_learning_ulmfit
jannenev/ulmfit-language-model pytorch
joelweber97/Python3_TF_Certificate tf
keithRebello/ULMFiT_sentiment_analysis
khumbuai/keras_wiki_lm tf
lukashedegaard/ride pytorch
lukexyz/Language-Models pytorch
magic-lantern/nlp-transfer-learning pytorch
mamamot/Russian-ULMFit pytorch
mdaniluk/language-detector
mrdbourke/tensorflow-deep-learning tf
muellerzr/CodeFest_2019
neburseni/NLP
nextbigwhat-ai/sentiment-analysis-pytorch pytorch
nishee0521/Sarcasm-Detector
noise-field/Russian-ULMFit pytorch
prajjwal1/language-modelling pytorch
radoslawkrolikowski/sentiment-analysis-pytorch pytorch
rajs96/ULMFiT-Twitter-US-Airline-Sentiment
rania000/SentAnalyser pytorch
rodrigopivi/aida tf
shivam360d/Sentiment-Analysis-ULMFit
simecek/Czech-ULMFiT
tanmaylaud/Patient_Conversation_Classifier_FastAI
tanvir-ishraq/healifyai--llm-based-healthcare-system pytorch
thisiskhan/tensorflow-developer-certificate-machine-learning-kit tf
uchange/ulangel pytorch
varshinireddyt/ULMFiT

Tasks

General ClassificationLanguage ModelingLanguage ModellingSentiment AnalysisText ClassificationTransfer Learning

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Adam 설명 없음
Sigmoid Activation 설명 없음
Tanh Activation 설명 없음
Temporal Activation Regularization 설명 없음
DropConnect DropConnect generalizes Dropout by randomly dropping the weights rather than the activations with probability $1-p$. DropConnect…
LSTM An LSTM is a type of recurrent neural network that addresses the vanishing gradient problem in vanilla…
Activation Regularization Activation Regularization (AR), or $L\_{2}$ activation regularization, is regularization performed on activations as opposed to weights. It is usually used in conjunction with…

Similar Papers 제목 키워드 기반

Building Efficient Universal Classifiers with Natural Language Inference

2023-12-29 · Moritz Laurer, Wouter van Atteveldt, Andreu Casas, Kasper Welbers

Generative Large Language Models (LLMs) have become the mainstream choice for fewshot and zeroshot learning thanks to the universality of text generation. Many users, however, do not need the broad capabilities of genera…

ClassificationNatural Language Inferencetext-classificationText Classification+1

How to Fine-Tune BERT for Text Classification?

2019-05-14 · Chi Sun, Xipeng Qiu, Yige Xu, Xuanjing Huang

Language model pre-training has proven to be useful in learning universal language representations. As a state-of-the-art language model pre-training model, BERT (Bidirectional Encoder Representations from Transformers) …

General ClassificationLanguage ModelingLanguage ModellingSentiment Analysis+1

Universal Language Model Fine-tuning for Patent Classification

2018-12-01 · ALTA 2018 12 · Jason Hepburn

This paper describes the methods used for the 2018 ALTA Shared Task. The task this year was to automatically classify Australian patents into their main International Patent Classification section. Our final submission u…

ClassificationGeneral ClassificationLanguage ModelingLanguage Modelling+3

UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment

2025-06-02 · Joseph Marvin Imperial, Abdullah Barayan, Regina Stodden, Rodrigo Wilkens 외

We introduce UniversalCEFR, a large-scale multilingual multidimensional dataset of texts annotated according to the CEFR (Common European Framework of Reference) scale in 13 languages. To enable open research in both aut…

P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks

2021-10-14 · Xiao Liu, Kaixuan Ji, Yicheng Fu, Weng Lam Tam 외

Prompt tuning, which only tunes continuous prompts with a frozen language model, substantially reduces per-task storage and memory usage at training. However, in the context of NLU, prior work reveals that prompt tuning …

Language ModelingLanguage Modelling