Search Results for author:
Found 59 papers, 34 papers with code
Why Knowledge Distillation Amplifies Gender Bias and How to Mitigate from the Perspective of DistilBERT
Knowledge distillation is widely used to transfer the language understanding of a large model to a smaller model. However, after knowledge distillation, it was found that the smaller model is more biased by gender compared to the source large model. This paper studies what causes gender bias to increase after the knowledge distillation process. Moreover, we suggest applying a variant of the mixup on knowledge distillation, which is used to increase generalizability during the distillation process, not for augmentation. By doing so, we can significantly reduce the gender bias amplification after knowledge distillation. We also conduct an experiment on the GLUE benchmark to demonstrate that even if the mixup is applied, it does not have a significant adverse effect on the model’s performance.
Suicidal Risk Detection for Military Personnel
We analyze social media for detecting the suicidal risk of military personnel, which is especially crucial for countries with compulsory military service such as the Republic of Korea.
Virtual Knowledge Graph Construction for Zero-Shot Domain-Specific Document Retrieval
We conduct three experiments 1) domain-specific document retrieval, 2) comparison of our virtual knowledge graph construction method with previous approaches, and 3) ablation study on each component of our virtual knowledge graph.
Perceptions to Beliefs: Exploring Precursory Inferences for Theory of Mind in Large Language Models
Our evaluation of eight state-of-the-art LLMs reveals that the models generally perform well in perception inference while exhibiting limited capability in perception-to-belief inference (e. g., lack of inhibitory control).
BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages
To address this issue, we introduce BLEnD, a hand-crafted benchmark designed to evaluate LLMs' everyday knowledge across diverse cultures and languages.
BEnQA: A Question Answering and Reasoning Benchmark for Bengali and English
In this study, we introduce BEnQA, a dataset comprising parallel Bengali and English exam questions for middle and high school levels in Bangladesh.
RECIPE4U: Student-ChatGPT Interaction Dataset in EFL Writing Education
RECIPE4U includes comprehensive records of these interactions, including conversation logs, students' intent, students' self-rated satisfaction, and students' essay edit histories.
CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Korean
Despite the rapid development of large language models (LLMs) for the Korean language, there remains an obvious lack of benchmark datasets that test the requisite Korean cultural and linguistic knowledge.
Multi-FAct: Assessing Multilingual LLMs' Multi-Regional Knowledge using FActScore
Large Language Models (LLMs) are prone to factuality hallucination, generating text that contradicts established knowledge.
Can LLM Generate Culturally Relevant Commonsense QA Data? Case Study in Indonesian and Sundanese
Large Language Models (LLMs) are increasingly being used to generate synthetic data for training and evaluating models.
The Generative AI Paradox on Evaluation: What It Can Solve, It May Not Evaluate
This paper explores the assumption that Large Language Models (LLMs) skilled in generation tasks are equally adept as evaluators.
Time-Aware Representation Learning for Time-Sensitive Question Answering
The model is trained to extract the answer span from the sentence that is both correct in time and context.
LLM-as-a-tutor in EFL Writing Education: Focusing on Evaluation of Student-LLM Interaction
In the context of English as a Foreign Language (EFL) writing education, LLM-as-a-tutor can assist students by providing real-time feedback on their essays.
ChEDDAR: Student-ChatGPT Dialogue in EFL Writing Education
We analyze students' usage patterns and perceptions regarding generative AI with respect to their intent and satisfaction.
Learning from Teaching Assistants to Program with Subgoals: Exploring the Potential for AI Teaching Assistants
To compare the learners' interaction and perception of the AI and human TAs, we conducted a between-subject study with 20 novice programming learners.
Exploring Cross-Cultural Differences in English Hate Speech Annotations: From Dataset Construction to Analysis
To address this, we introduce CREHate, a CRoss-cultural English Hate speech dataset.
KoBBQ: Korean Bias Benchmark for Question Answering
In this paper, we present KoBBQ, a Korean bias benchmark dataset, and we propose a general framework that addresses considerations for cultural adaptation of a dataset.
SQuARe: A Large-Scale Dataset of Sensitive Questions and Acceptable Responses Created Through Human-Machine Collaboration
The potential social harms that large language models pose, such as generating offensive content and reinforcing biases, are steeply rising.
CS1QA: A Dataset for Assisting Code-based Question Answering in an Introductory Programming Course
We introduce CS1QA, a dataset for code-based question answering in the programming education domain.
IDK-MRC: Unanswerable Questions for Indonesian Machine Reading Comprehension
Machine Reading Comprehension (MRC) has become one of the essential tasks in Natural Language Understanding (NLU) as it is often included in several NLU benchmarks (Liang et al., 2020; Wilie et al., 2020).
Machine Reading Comprehension
Natural Language Understanding
+1
Towards standardizing Korean Grammatical Error Correction: Datasets and Annotation
We show that the model trained with our datasets significantly outperforms the currently used statistical Korean GEC system (Hanspell) on a wider range of error types, demonstrating the diversity and usefulness of the datasets.
Rethinking Annotation: Can Language Learners Contribute?
Researchers have traditionally recruited native speakers to provide annotations for widely used benchmark datasets.
HUE: Pretrained Model and Dataset for Understanding Hanja Documents of Ancient Korea
Historical records in Korea before the 20th century were primarily written in Hanja, an extinct language based on Chinese characters and not understood by modern Korean or Chinese speakers.
Ranking-Enhanced Unsupervised Sentence Representation Learning
In this work, we show that the semantic meaning of a sentence is also determined by nearest-neighbor sentences that are similar to the input sentence.
Models and Benchmarks for Representation Learning of Partially Observed Subgraphs
Subgraphs are rich substructures in graphs, and their nodes and edges can be partially observed in real-world tasks.
KOLD: Korean Offensive Language Dataset
Recent directions for offensive language detection are hierarchical modeling, identifying the type and the target of offensive language, and interpretability with offensive span annotation and prediction.
Translating Hanja Historical Documents to Contemporary Korean and English
Built on top of multilingual neural machine translation, H2KE learns to translate a historical document written in Hanja, from both a full dataset of outdated Korean translation and a small dataset of more recently translated contemporary Korean and English.
Two-Step Question Retrieval for Open-Domain QA
These models have shown a significant increase in inference speed, but at the cost of lower QA performance compared to the retriever-reader models.
How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision
However, what graph attention learns is not understood well, particularly when graphs are noisy.
Translating Subgraphs to Nodes Makes Simple GNNs Strong and Efficient for Subgraph Representation Learning
Subgraph representation learning has emerged as an important problem, but it is by default approached with specialized graph neural networks on a large global graph.
Emergent Communication under Varying Sizes and Connectivities
Just as we humans have succeeded in creating a shared language that allows us to interact within a large group, can the emergent communication within an artificial group converge to a shared, agreed language?
Learning Representations of Partial Subgraphs by Subgraph InfoMax
Subgraphs are important substructures of graphs, but learning their representations has not been studied well.
Knowledge-Enhanced Evidence Retrieval for Counterargument Generation
Finding counterevidence to statements is key to many tasks, including counterargument generation.
Learning Bill Similarity with Annotated and Augmented Corpora of Bills
Despite the significance of bill-to-bill linkages for understanding the legislative process, existing approaches fail to address semantic similarities across bills, let alone reordering or paraphrasing which are prevalent in legal document writing.
Mitigating Language-Dependent Ethnic Bias in BERT
Which of the two methods works better depends on the amount of NLP resources available for that language.
Efficient Contrastive Learning via Novel Data Augmentation and Curriculum Learning
We introduce EfficientCL, a memory-efficient continual pretraining method that applies contrastive learning with novel data augmentation and curriculum learning.
Weakly Supervised Pre-Training for Multi-Hop Retriever
In multi-hop QA, answering complex questions entails iterative document retrieval for finding the missing entity of the question.
KLUE: Korean Language Understanding Evaluation
We introduce Korean Language Understanding Evaluation (KLUE) benchmark.
Context-Aware Answer Extraction in Question Answering
With experiments on reading comprehension, we show that BLANC outperforms the state-of-the-art QA models, and the performance gap increases as the number of answer text occurrences increases.
Speaker Sensitive Response Evaluation Model
We provide our code and the learned parameters so that they can be used for automatic evaluation of dialogue response generation models.
K-EmoCon, a multimodal sensor dataset for continuous emotion recognition in naturalistic conversations
Therefore, studying emotions in the context of social interactions requires a novel dataset, and K-EmoCon is such a multimodal dataset with comprehensive annotations of continuous emotions during naturalistic conversations.
Dimensional Emotion Detection from Categorical Emotion
We present a model to predict fine-grained emotions along the continuous dimensions of valence, arousal, and dominance (VAD) with a corpus with categorical emotion annotations.
Variational Hierarchical User-based Conversation Model
To overcome this limitation, we propose a new model with a stochastic variable designed to capture the speaker information and deliver it to the conversational context.
Additive Compositionality of Word Vectors
Additive compositionality of word embedding models has been studied from empirical and theoretical perspectives.
Emergence of Collective Policies Inside Simulations with Biased Representations
We consider a setting where biases are involved when agents internalise an environment.
Conversation Model Fine-Tuning for Classifying Client Utterances in Counseling Dialogues
A dataset of those interactions can be used to learn to automatically classify the client utterances into categories that help counselors in diagnosing client status and predicting counseling outcome.
Homogeneity-Based Transmissive Process to Model True and False News in Social Networks
An overwhelming number of true and false news stories are posted and shared in social networks, and users diffuse the stories based on multiple factors.
Hierarchical Dirichlet Gaussian Marked Hawkes Process for Narrative Reconstruction in Continuous Time Domain
In news and discussions, many articles and posts are provided without their related previous articles or posts.
Conversational Decision-Making Model for Predicting the King's Decision in the Annals of the Joseon Dynasty
Styles of leaders when they make decisions in groups vary, and the different styles affect the performance of the group.
Subword-level Word Vector Representations for Korean
The results show that our simple method outperforms word2vec and character-level Skip-Grams on semantic and syntactic similarity and analogy tasks and contributes positively toward downstream NLP tasks such as sentiment analysis.
Leveraging the Crowd to Detect and Reduce the Spread of Fake News and Misinformation
Online social networking sites are experimenting with the following crowd-powered procedure to reduce the spread of fake news and misinformation: whenever a user is exposed to a story through her feed, she can flag the story as misinformation and, if the story receives enough flags, it is sent to a trusted third party for fact checking.
Rotated Word Vector Representations and their Interpretability
We apply several rotation algorithms to the vector representation of words to improve the interpretability.
Joint Modeling of Topics, Citations, and Topical Authority in Academic Corpora
Second, it models each author's influence on citations of a paper based on the topics of the cited papers, as well as the citing papers.
Understanding Editing Behaviors in Multilingual Wikipedia
We study multilingualism by collecting and analyzing a large dataset of the content written by multilingual editors of the English, German, and Spanish editions of Wikipedia.
Hierarchical Dirichlet Scaling Process
We present the \textit{hierarchical Dirichlet scaling process} (HDSP), a Bayesian nonparametric mixed membership model.
