Ho-Jin Choi

Also published as: Ho-jin Choi


2024

pdf bib
Large Language Models can Share Images, Too!
Young-Jun Lee | Dokyong Lee | Joo Won Sung | Jonghwan Hyeon | Ho-Jin Choi
Findings of the Association for Computational Linguistics ACL 2024

This paper explores the image-sharing capability of Large Language Models (LLMs), such as GPT-4 and LLaMA 2, in a zero-shot setting. To facilitate a comprehensive evaluation of LLMs, we introduce the photochatplus dataset, which includes enriched annotations (ie intent, triggering sentence, image description, and salient information). Furthermore, we present the gradient-free and extensible Decide, Describe, and Retrieve () framework. With extensive experiments, we unlock the image-sharing capability of equipped with LLMs in zero-shot prompting, with ChatGPT achieving the best performance.Our findings also reveal the emergent image-sharing ability in LLMs under zero-shot conditions, validating the effectiveness of . We use this framework to demonstrate its practicality and effectiveness in two real-world scenarios: (1) human-bot interaction and (2) dataset augmentation. To the best of our knowledge, this is the first study to assess the image-sharing ability of various LLMs in a zero-shot setting. We make our source code and dataset publicly available at https://github.com/passing2961/DribeR.

pdf bib
DialogCC: An Automated Pipeline for Creating High-Quality Multi-Modal Dialogue Dataset
Young-Jun Lee | Byungsoo Ko | Han-Gyu Kim | Jonghwan Hyeon | Ho-Jin Choi
Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)

As sharing images in an instant message is a crucial factor, there has been active research on learning an image-text multi-modal dialogue models.However, training a well-generalized multi-modal dialogue model remains challenging due to the low quality and limited diversity of images per dialogue in existing multi-modal dialogue datasets.In this paper, we propose an automated pipeline to construct a multi-modal dialogue dataset, ensuring both dialogue quality and image diversity without requiring minimum human effort. In our pipeline, to guarantee the coherence between images and dialogue, we prompt GPT-4 to infer potential image-sharing moments - specifically, the utterance, speaker, rationale, and image description. Furthermore, we leverage CLIP similarity to maintain consistency between aligned multiple images to the utterance.Through this pipeline, we introduce DialogCC, a high-quality and diverse multi-modal dialogue dataset that surpasses existing datasets in terms of quality and diversity in human evaluation.Our comprehensive experiments highlight that when multi-modal dialogue models are trained using our dataset, their generalization performance on unseen dialogue datasets is significantly enhanced. We make our source code and dataset publicly available (https://dialogcc.github.io/).

2023

pdf bib
Semantic Ambiguity Detection in Sentence Classification using Task-Specific Embeddings
Jong Myoung Kim | Young-jun Lee | Sangkeun Jung | Ho-jin Choi
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)

Ambiguity is a major obstacle to providing services based on sentence classification. However, because of the structural limitations of the service, there may not be sufficient contextual information to resolve the ambiguity. In this situation, we focus on ambiguity detection so that service design considering ambiguity is possible. We utilize similarity in a semantic space to detect ambiguity in service scenarios and training data. In addition, we apply task-specific embedding to improve performance. Our results demonstrate that ambiguities and resulting labeling errors in training data or scenarios can be detected. Additionally, we confirm that it can be used to debug services

2022

pdf bib
Pneg: Prompt-based Negative Response Generation for Dialogue Response Selection Task
Nyoungwoo Lee | ChaeHun Park | Ho-Jin Choi | Jaegul Choo
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

In retrieval-based dialogue systems, a response selection model acts as a ranker to select the most appropriate response among several candidates. However, such selection models tend to rely on context-response content similarity, which makes models vulnerable to adversarial responses that are semantically similar but not relevant to the dialogue context. Recent studies have shown that leveraging these adversarial responses as negative training samples is useful for improving the discriminating power of the selection model. Nevertheless, collecting human-written adversarial responses is expensive, and existing synthesizing methods often have limited scalability. To overcome these limitations, this paper proposes a simple but efficient method for generating adversarial negative responses leveraging a large-scale language model. Experimental results on dialogue selection tasks show that our method outperforms other methods of synthesizing adversarial negative responses. These results suggest that our method can be an effective alternative to human annotators in generating adversarial responses. Our code and dataset will be released if the paper is accepted.

pdf bib
Does GPT-3 Generate Empathetic Dialogues? A Novel In-Context Example Selection Method and Automatic Evaluation Metric for Empathetic Dialogue Generation
Young-Jun Lee | Chae-Gyun Lim | Ho-Jin Choi
Proceedings of the 29th International Conference on Computational Linguistics

Since empathy plays a crucial role in increasing social bonding between people, many studies have designed their own dialogue agents to be empathetic using the well-established method of fine-tuning. However, they do not use prompt-based in-context learning, which has shown powerful performance in various natural language processing (NLP) tasks, for empathetic dialogue generation. Although several studies have investigated few-shot in-context learning for empathetic dialogue generation, an in-depth analysis of the generation of empathetic dialogue with in-context learning remains unclear, especially in GPT-3 (Brown et al., 2020). In this study, we explore whether GPT-3 can generate empathetic dialogues through prompt-based in-context learning in both zero-shot and few-shot settings. To enhance performance, we propose two new in-context example selection methods, called SITSM and EMOSITSM, that utilize emotion and situational information. We also introduce a new automatic evaluation method, DIFF-EPITOME, which reflects the human tendency to express empathy. From the analysis, we reveal that our DIFF-EPITOME is effective in measuring the degree of human empathy. We show that GPT-3 achieves competitive performance with Blender 90M, a state-of-the-art dialogue generative model, on both automatic and human evaluation. Our code is available at https://github.com/passing2961/EmpGPT-3.

pdf bib
PERSONACHATGEN: Generating Personalized Dialogues using GPT-3
Young-Jun Lee | Chae-Gyun Lim | Yunsu Choi | Ji-Hui Lm | Ho-Jin Choi
Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge

Recently, many prior works have made their own agents generate more personalized and engaging responses using personachat. However, since this dataset is frozen in 2018, the dialogue agents trained on this dataset would not know how to interact with a human who loves “Wandavision.” One way to alleviate this problem is to create a large-scale dataset. In this work, we introduce the pipeline of creating personachatgen, which is comprised of three main components: Creating (1) profilegen, (2) Persona Set, and (3) personachatgen. To encourage GPT-3’s generation ability, we also defined a taxonomy of hierarchical persona category derived from social profiling taxonomy. To create the speaker consistent persona set, we propose a simple contradiction-based iterative sentence replacement algorithm, named CoNL. Moreover, to prevent GPT-3 generating harmful content, we presented two filtering pipelines, one each for profilegen and personachatgen. Through analyzing of personachatgen, we showed that GPT-3 can generate personalized dialogue containing diverse persona. Furthermore, we revealed a state-of-the-art Blender 90M trained on our dataset that leads to higher performance.

2021

pdf bib
Constructing Multi-Modal Dialogue Dataset by Replacing Text with Semantically Relevant Images
Nyoungwoo Lee | Suwon Shin | Jaegul Choo | Ho-Jin Choi | Sung-Hyon Myaeng
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)

In multi-modal dialogue systems, it is important to allow the use of images as part of a multi-turn conversation. Training such dialogue systems generally requires a large-scale dataset consisting of multi-turn dialogues that involve images, but such datasets rarely exist. In response, this paper proposes a 45k multi-modal dialogue dataset created with minimal human intervention. Our method to create such a dataset consists of (1) preparing and pre-processing text dialogue datasets, (2) creating image-mixed dialogues by using a text-to-image replacement technique, and (3) employing a contextual-similarity-based filtering step to ensure the contextual coherence of the dataset. To evaluate the validity of our dataset, we devise a simple retrieval model for dialogue sentence prediction tasks. Automatic metrics and human evaluation results on such tasks show that our dataset can be effectively used as training data for multi-modal dialogue systems which require an understanding of images and text in a context-aware manner. Our dataset and generation code is available at https://github.com/shh1574/multi-modal-dialogue-dataset.

2020

pdf bib
Korean-Specific Emotion Annotation Procedure Using N-Gram-Based Distant Supervision and Korean-Specific-Feature-Based Distant Supervision
Young-Jun Lee | Chae-Gyun Lim | Ho-Jin Choi
Proceedings of the Twelfth Language Resources and Evaluation Conference

Detecting emotions from texts is considerably important in an NLP task, but it has the limitation of the scarcity of manually labeled data. To overcome this limitation, many researchers have annotated unlabeled data with certain frequently used annotation procedures. However, most of these studies are focused mainly on English and do not consider the characteristics of the Korean language. In this paper, we present a Korean-specific annotation procedure, which consists of two parts, namely n-gram-based distant supervision and Korean-specific-feature-based distant supervision. We leverage the distant supervision with the n-gram and Korean emotion lexicons. Then, we consider the Korean-specific emotion features. Through experiments, we showed the effectiveness of our procedure by comparing with the KTEA dataset. Additionally, we constructed a large-scale emotion-labeled dataset, Korean Movie Review Emotion (KMRE) Dataset, using our procedure. In order to construct our dataset, we used a large-scale sentiment movie review corpus as the unlabeled dataset. Moreover, we used a Korean emotion lexicon provided by KTEA. We also performed an emotion classification task and a human evaluation on the KMRE dataset.

2018

pdf bib
Korean TimeBank Including Relative Temporal Information
Chae-Gyun Lim | Young-Seob Jeong | Ho-Jin Choi
Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)

2016

pdf bib
Korean TimeML and Korean TimeBank
Young-Seob Jeong | Won-Tae Joo | Hyun-Woo Do | Chae-Gyun Lim | Key-Sun Choi | Ho-Jin Choi
Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16)

Many emerging documents usually contain temporal information. Because the temporal information is useful for various applications, it became important to develop a system of extracting the temporal information from the documents. Before developing the system, it first necessary to define or design the structure of temporal information. In other words, it is necessary to design a language which defines how to annotate the temporal information. There have been some studies about the annotation languages, but most of them was applicable to only a specific target language (e.g., English). Thus, it is necessary to design an individual annotation language for each language. In this paper, we propose a revised version of Koreain Time Mark-up Language (K-TimeML), and also introduce a dataset, named Korean TimeBank, that is constructed basd on the K-TimeML. We believe that the new K-TimeML and Korean TimeBank will be used in many further researches about extraction of temporal information.

2015

pdf bib
Temporal Information Extraction from Korean Texts
Young-Seob Jeong | Zae Myung Kim | Hyun-Woo Do | Chae-Gyun Lim | Ho-Jin Choi
Proceedings of the Nineteenth Conference on Computational Natural Language Learning

pdf bib
Korean Twitter Emotion Classification Using Automatically Built Emotion Lexicons and Fine-Grained Features
Hyo Jin Do | Ho-Jin Choi
Proceedings of the 29th Pacific Asia Conference on Language, Information and Computation: Posters

pdf bib
Measuring Popularity of Machine-Generated Sentences Using Term Count, Document Frequency, and Dependency Language Model
Jong Myoung Kim | Hancheol Park | Young-Seob Jeong | Ho-Jin Choi | Gahgene Gweon | Jeong Hur
Proceedings of the 29th Pacific Asia Conference on Language, Information and Computation: Posters

2014

pdf bib
Sentential Paraphrase Generation for Agglutinative Languages Using SVM with a String Kernel
Hancheol Park | Gahgene Gweon | Ho-Jin Choi | Jeong Heo | Pum-Mo Ryu
Proceedings of the 28th Pacific Asia Conference on Language, Information and Computing