• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.524-527
• /
• 2021

최근 국외에서 사실 검증 연구가 활발하게 이루어지고 있지만 한국어의 경우 데이터 집합의 부재로 인하여 사실 검증 연구가 이루어지는데 큰 어려움을 겪고 있다. 이러한 어려움을 해소하고자 자동 생성 모델을 통하여 데이터 집합을 생성하는 시도도 있으나 생성 모델의 특성 상 부정확한 데이터가 생성되어 사실 검증 연구의 퀄리티를 떨어뜨린다는 문제점이 있다. 이러한 문제점을 해소하기 위해 수동으로 구축한 100건의 데이터 집합으로 최근에 이루어진 퓨-샷(Few-Shot) 사실 검증을 확장한 학습이 필요없는 제로-샷(Zero-Shot) 질의 응답에 대한 사실 검증 연구를 제안한다.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.43-50
• /
• 2024

This paper evaluates the Korean sentiment analysis performance of large language models like GPT-3.5 and GPT-4 using a zero-shot approach facilitated by the ChatGPT API, comparing them to pre-trained Korean models such as KoBERT. Through experiments utilizing various Korean sentiment analysis datasets in fields like movies, gaming, and shopping, the efficiency of these models is validated. The results reveal that the LMKor-ELECTRA model displayed the highest performance based on F1-score, while GPT-4 particularly achieved high accuracy and F1-scores in movie and shopping datasets. This indicates that large language models can perform effectively in Korean sentiment analysis without prior training on specific datasets, suggesting their potential in zero-shot learning. However, relatively lower performance in some datasets highlights the limitations of the zero-shot based methodology. This study explores the feasibility of using large language models for Korean sentiment analysis, providing significant implications for future research in this area.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.35-43
• /
• 2021

Recently, there has been a growing interest in zero-shot cross-lingual transfer, which leverages cross-lingual language models (CLLMs) to perform downstream tasks that are not trained in a specific language. In this paper, we point out the limitations of the data-centric aspect of quality estimation (QE), and perform zero-shot cross-lingual transfer even in environments where it is difficult to construct QE data. Few studies have dealt with zero-shots in QE, and after fine-tuning the English-German QE dataset, we perform zero-shot transfer leveraging CLLMs. We conduct comparative analysis between various CLLMs. We also perform zero-shot transfer on language pairs with different sized resources and analyze results based on the linguistic characteristics of each language. Experimental results showed the highest performance in multilingual BART and multillingual BERT, and we induced QE to be performed even when QE learning for a specific language pair was not performed at all.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.69-74
• /
• 2023

This study introduces an innovative model for zero-shot voice conversion that utilizes the capabilities of HuBERT. Zero-shot voice conversion models can transform the speech of one speaker to mimic that of another, even when the model has not been exposed to the target speaker's voice during the training phase. Comprising five main components (HuBERT, feature encoder, flow, speaker encoder, and vocoder), the model offers remarkable performance across a range of scenarios. Notably, it excels in the challenging unseen-to-unseen voice-conversion tasks. The effectiveness of the model was assessed based on the mean opinion scores and similarity scores, reflecting high voice quality and similarity to the target speakers. This model demonstrates considerable promise for a range of real-world applications demanding high-quality voice conversion. This study sets a precedent in the exploration of HuBERT-based models for voice conversion, and presents new directions for future research in this domain. Despite its complexities, the robust performance of this model underscores the viability of HuBERT in advancing voice conversion technology, making it a significant contributor to the field.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.399-403
• /
• 2023

본 논문에서는 다중 문서 기반의 대화 시스템을 통한 효율적인 정보 검색과 응답 생성에 중점을 둡니다. 대규모 데이터 집합에서 정확한 문서를 선택하는 데 필요한 검색의 중요성을 강조하며, 현재 검색 방법의 한계와 문제점을 지적합니다. 또한 더 자연스러운 답변을 생성하기 위해 대규모 언어 모델을 사용하게 되면서 fine-tuning 시에 발생하는 제약과 낭비를 모델의 제로샷 생성 능력을 활용하여 개선하려는 방안을 제안하며, 모델의 크기와 자원의 효율성에 대한 고려사항을 논의합니다. 우리의 접근 방식은 대규모 언어 모델을 프롬프트와 함께 다중 문서로 학습 없이 정보를 검색하고 응답을 생성하는 방향으로 접근하여 대화 시스템의 효율성과 유용성을 향상시킬 수 있음을 제시합니다.

• Journal of the Korean Statistical Society
• /
• v.7 no.1
• /
• pp.61-80
• /
• 1978

This paper presents general solutions for stochastic square duels with continuous interfiring times and various firing strategies such as standby (S), concentrated (C), seperated (I) and random (R) firings. Analysis of these square duels with negative exponential interfiring times and equivalent values of rates of fire and single shot kill probabilities reveal three important facts: i) Strategy (C) is advantageous against the opponent's strategy (S) and the advantage becomes more pronounced for lower values of single shot kill probabilities. ii) Strategy (I) is always better than strategy (C) no matter which of (C) and (I) the opponent uses and its relative advantege increases to a quarter as single shot kill probabilities increase to one but decreases to zero as they go to zero. iii) However, strategy (I) has no advantage over strategy (C) for small values of single shot kill probabilities. In this paper, square duels with strategies (C) and (I) are based on the assumptions that duelists are homogeneous and both duelists of one side fire simultaneously. The problem of relaxing these assumptions and extension of square ($2 \times 2$) duels to more general ($m \times n) duels are now being investigated.

• Journal of Sensor Science and Technology
• /
• v.9 no.4
• /
• pp.316-326
• /
• 2000

A new signal processing algorithm for white light interferometry has been proposed and investigated theoretically. The goal of the algorithm is to determine the absolute optical path length of an interferometer with very high precision (<< one optical wavelength). The algorithm features cross-correlation of interferometer fringe scans and hypothesis testing. The hypothesis test looks for a zero order fringe peak candidate about which the cross-correlation is symmetric minimizing the uncertainty of misidentification. The shot noise limited performance of the proposed signal processing algorithm has been analyzed using computer simulations. Simulation results were extrapolated to predict the misidentification rate at Signal to-Shot noise ratio (SNR) higher than 31 dB. Root-mean-square phase error between the computer-generated zero order fringe peak and the estimated zero order fringe peak has been calculated for the changes of three different parameters (SNR, fringe scan sampling rate, coherence length of light source). Results of computer simulations showed the ability of the proposed signal processing algorithm to identify the zero order fringe peak correctly. The proposed signal processing algorithm uses a software approach, which is potentially inexpensive, simple and fast.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.473-475
• /
• 2021

유의어 추천을 구현하기 위해서는 각 단어 사이의 유사도를 계산하는 것이 필수적이다. 하지만, 기존의 단어간 유사도를 계산하는 여러 방법들은 데이터셋에 등장하지 않은 단어에 대해 유사도를 계산 할 수 없다. 이 논문에서는 이를 해결하기 위해 언어모델의 PPL을 활용하여 단어간 유사도를 계산하였고, 이를 통해 유의어를 추천했을 때 MRR 41.31%의 성능을 확인했다.

• Smart Structures and Systems
• /
• v.33 no.1
• /
• pp.55-64
• /
• 2024

This paper proposes a fusion imaging-based coating-defect classification method for steel structures that uses zero-shot learning. In the proposed method, a halogen lamp generates heat energy on the coating surface of a steel structure, and the resulting heat responses are measured by an infrared (IR) camera, while photos of the coating surface are captured by a charge-coupled device (CCD) camera. The measured heat responses and visual images are then analyzed using zero-shot learning to classify the coating defects, and the estimated coating defects are visualized throughout the inspection surface of the steel structure. In contrast to older approaches to coating-defect classification that relied on visual inspection and were limited to surface defects, and older artificial neural network (ANN)-based methods that required large amounts of data for training and validation, the proposed method accurately classifies both internal and external defects and can classify coating defects for unobserved classes that are not included in the training. Additionally, the proposed model easily learns about additional classifying conditions, making it simple to add classes for problems of interest and field application. Based on the results of validation via field testing, the defect-type classification performance is improved 22.7% of accuracy by fusing visual and thermal imaging compared to using only a visual dataset. Furthermore, the classification accuracy of the proposed method on a test dataset with only trained classes is validated to be 100%. With word-embedding vectors for the labels of untrained classes, the classification accuracy of the proposed method is 86.4%.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.722-725
• /
• 2023

대규모 언어 모델(LLM)은 대규모의 데이터를 학습하여 얻은 지식을 기반으로 텍스트와 다양한 콘텐츠를 인식하고 요약, 번역, 예측, 생성할 수 있는 딥러닝 알고리즘이다. 초기 공개된 LLM은 영어 기반 모델로 비영어권에서는 높은 성능을 기대할 수 없었으며, 이에 한국, 중국 등 자체적 LLM 연구개발이 활성화되고 있다. 본 논문에서는 언어가 LLM의 성능에 영향을 미치는가에 대하여 한국어 기반 LLM과 영어 기반 LLM으로 KoBEST의 4가지 Task에 대하여 성능비교를 하였다. 그 결과 한국어에 대한 사전 지식을 추가하는 것이 LLM의 성능에 영향을 미치는 것을 확인할 수 있었다.