Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userRedwoodNLP at SemEval-2021 Task 7: Ensembled Pretrained and Lightweight Models for Humor Detection
redwoodnlp في مهمة Semeval-2021 7: نموذج نماذج محددة وخفيفة الوزن للكشف عن فكاهة
253
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
إن فهم الفكاهة هو عنصر أساسي لأنظمة NLP التي تواجه الإنسان.في هذه الورقة، نحقق في العديد من الطرق للكشف عن الفكاهة في تصريحات قصيرة كجزء من المهمة المشتركة SEMEVAL-2021 7. للمهمة 1A، نطبق مجموعة من نماذج اللغة المدربة مسبقا مسبقا؛بالنسبة للمهام 1B، 1C، و 2A، نحقق في العديد من نماذج التعلم الآلية القائمة على الأشجار والخطية.ينص نظامنا النهائي على درجة F1 من 0.9571 (المرتبة 24/58) في المهمة 1A، ورمز من 0.5580 (مرتبة 18/50) في المهمة 1B، درجة F1 من 0.5024 (المرتبة 26/36) في مهمة 1C،ورمز من 0.7229 (المرتبة 45/88) في مهمة 2A.
An understanding of humor is an essential component of human-facing NLP systems. In this paper, we investigate several methods for detecting humor in short statements as part of Semeval-2021 Shared Task 7. For Task 1a, we apply an ensemble of fine-tuned pre-trained language models; for Tasks 1b, 1c, and 2a, we investigate various tree-based and linear machine learning models. Our final system achieves an F1-score of 0.9571 (ranked 24 / 58) on Task 1a, an RMSE of 0.5580 (ranked 18 / 50) on Task 1b, an F1-score of 0.5024 (ranked 26 / 36) on Task 1c, and an RMSE of 0.7229 (ranked 45 / 48) on Task 2a.
References used
https://aclanthology.org/
rate research
Read More
This paper presents the DuluthNLP submission to Task 7 of the SemEval 2021 competition on Detecting and Rating Humor and Offense. In it, we explain the approach used to train the model together with the process of fine-tuning our model in getting the
results. We focus on humor detection, rating, and of-fense rating, representing three out of the four subtasks that were provided. We show that optimizing hyper-parameters for learning rate, batch size and number of epochs can increase the accuracy and F1 score for humor detection
Humor detection has become a topic of interest for several research teams, especially those involved in socio-psychological studies, with the aim to detect the humor and the temper of a targeted population (e.g. a community, a city, a country, the em
ployees of a given company). Most of the existing studies have formulated the humor detection problem as a binary classification task, whereas it revolves around learning the sense of humor by evaluating its different degrees. In this paper, we propose an end-to-end deep Multi-Task Learning (MTL) model to detect and rate humor and offense. It consists of a pre-trained transformer encoder and task-specific attention layers. The model is trained using MTL uncertainty loss weighting to adaptively combine all sub-tasks objective functions. Our MTL model tackles all sub-tasks of the SemEval-2021 Task-7 in one end-to-end deep learning system and shows very promising results.
This paper describes our contribution to SemEval-2021 Task 7: Detecting and Rating Humor and Of-fense.This task contains two sub-tasks, sub-task 1and sub-task 2. Among them, sub-task 1 containsthree sub-tasks, sub-task 1a ,sub-task 1b and sub-task 1c
.Sub-task 1a is to predict if the text would beconsidered humorous.Sub-task 1c is described asfollows: if the text is classed as humorous, predictif the humor rating would be considered controver-sial, i.e. the variance of the rating between annota-tors is higher than the median.we combined threepre-trained model with CNN to complete these twoclassification sub-tasks.Sub-task 1b is to judge thedegree of humor.Sub-task 2 aims to predict how of-fensive a text would be with values between 0 and5.We use the idea of regression to deal with thesetwo sub-tasks.We analyze the performance of ourmethod and demonstrate the contribution of eachcomponent of our architecture.We have achievedgood results under the combination of multiple pre-training models and optimization methods.
Humor detection and rating poses interesting linguistic challenges to NLP; it is highly subjective depending on the perceptions of a joke and the context in which it is used. This paper utilizes and compares transformers models; BERT base and Large,
BERTweet, RoBERTa base and Large, and RoBERTa base irony, for detecting and rating humor and offense. The proposed models, where given a text in cased and uncased type obtained from SemEval-2021 Task7: HaHackathon: Linking Humor and Offense Across Different Age Groups. The highest scored model for the first subtask: Humor Detection, is BERTweet base cased model with 0.9540 F1-score, for the second subtask: Average Humor Rating Score, it is BERT Large cased with the minimum RMSE of 0.5555, for the fourth subtask: Average Offensiveness Rating Score, it is BERTweet base cased model with minimum RMSE of 0.4822.
This paper describes our system participated in Task 7 of SemEval-2021: Detecting and Rating Humor and Offense. The task is designed to detect and score humor and offense which are influenced by subjective factors. In order to obtain semantic informa
tion from a large amount of unlabeled data, we applied unsupervised pre-trained language models. By conducting research and experiments, we found that the ERNIE 2.0 and DeBERTa pre-trained models achieved impressive performance in various subtasks. Therefore, we applied the above pre-trained models to fine-tune the downstream neural network. In the process of fine-tuning the model, we adopted multi-task training strategy and ensemble learning method. Based on the above strategy and method, we achieved RMSE of 0.4959 for subtask 1b, and finally won the first place.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
