Beomsu Kim
Research Subjects
Beomsu Kim has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Buru Chang
Korea University, Seoul, Republic of Korea
3 shared publications Enkhbayar Erdenee
2 shared publications- Seokjun Seo
Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea
2 shared publications - Seungju Han
Machine Learning TU, Samsung Advanced Institute of Technology, Suwon, Republic of Korea
2 shared publications Jin Yong Yoo
1 shared publicationJoonyoung Yi
Hyperconnect, South Korea
1 shared publication
Latest Publications
- Open Access
- 1 January 2022
- Published by Association for Computational Linguistics (ACL)
- p. 5114-5132
Meet Your Favorite Character: Open-domain Chatbot Mimicking Fictional Characters with only a Few Utterances
Seungju Han, Beomsu Kim, Jin Yong Yoo, Seokjun Seo, Sangbum Kim, Enkhbayar Erdenee, Buru Chang. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.
- 1 January 2022
- Published by Association for Computational Linguistics (ACL)
- p. 5114-5132
Open AccessMeet Your Favorite Character: Open-domain Chatbot Mimicking Fictional Characters with only a Few Utterances
Seungju Han, Beomsu Kim, Jin Yong Yoo, Seokjun Seo, Sangbum Kim, Enkhbayar Erdenee, Buru Chang. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.
- Open Access
- 1 January 2022
- Published by Association for Computational Linguistics (ACL)
- p. 218-230
Understanding and Improving the Exemplar-based Generation for Open-domain Conversation
Seungju Han, Beomsu Kim, Seokjun Seo, Enkhbayar Erdenee, Buru Chang. Proceedings of the 4th Workshop on NLP for Conversational AI. 2022.
- 1 January 2022
- Published by Association for Computational Linguistics (ACL)
- p. 218-230
Open AccessUnderstanding and Improving the Exemplar-based Generation for Open-domain Conversation
Seungju Han, Beomsu Kim, Seokjun Seo, Enkhbayar Erdenee, Buru Chang. Proceedings of the 4th Workshop on NLP for Conversational AI. 2022.
- 1 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 75-84
Embedding Normalization: Significance Preserving Feature Normalization for Click-Through Rate Prediction
Normalization techniques are known to provide faster model convergence and good generalization performance, having achieved great success in computer vision and natural language processing. Recently, several deep neural network-based click-through rate (CTR) prediction models have applied such normalization techniques to their deep network components to make model training stable. However, we observe that applying existing normalization techniques (e.g. Batch Normalization and Layer Normalization) to feature embedding of the models leads to the degradation of model performance. In this study, we conjecture that existing normalization techniques can easily ignore the significance of each feature embedding, leading to sub-optimal performance. To support our claim, we theoretically show that existing normalization techniques tend to equalize the norm of individual feature embedding. To overcome this limitation, we propose a theory-inspired normalization technique, called Embedding Normalization, which not only makes model training stable but also improves the performance of CTR prediction models by preserving the significance of each feature embedding. Through extensive experiments on various real-world CTR prediction datasets, we show that our proposed normalization technique leads to faster model convergence and achieves better or comparable performance than other normalization techniques. Especially, our Embedding Normalization is effective in not only deep neural network-based CTR prediction models but also shallow CTR prediction models that do not utilize deep neural network components.
- 1 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 75-84
Embedding Normalization: Significance Preserving Feature Normalization for Click-Through Rate Prediction
Normalization techniques are known to provide faster model convergence and good generalization performance, having achieved great success in computer vision and natural language processing. Recently, several deep neural network-based click-through rate (CTR) prediction models have applied such normalization techniques to their deep network components to make model training stable. However, we observe that applying existing normalization techniques (e.g. Batch Normalization and Layer Normalization) to feature embedding of the models leads to the degradation of model performance. In this study, we conjecture that existing normalization techniques can easily ignore the significance of each feature embedding, leading to sub-optimal performance. To support our claim, we theoretically show that existing normalization techniques tend to equalize the norm of individual feature embedding. To overcome this limitation, we propose a theory-inspired normalization technique, called Embedding Normalization, which not only makes model training stable but also improves the performance of CTR prediction models by preserving the significance of each feature embedding. Through extensive experiments on various real-world CTR prediction datasets, we show that our proposed normalization technique leads to faster model convergence and achieves better or comparable performance than other normalization techniques. Especially, our Embedding Normalization is effective in not only deep neural network-based CTR prediction models but also shallow CTR prediction models that do not utilize deep neural network components.
- 11 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 21 (6) , 1345-1358
New Hole-Filling Method Using Extrapolated Spatio-Temporal Background Information for a Synthesized Free-View
This paper introduces a new hole-filling method using extrapolated spatio-temporal background information to obtain a synthesized free-view. New temporal background modeling is proposed, which incorporates stationary temporal information in the hole-filling process and preserves the temporal consistency of synthesized views. A background codebook is distinguished from a non-overlapped patch-based codebook, which contributes to extracting reliable temporal background information. Furthermore, a depth-map driven spatial local background estimation is also addressed to discriminate the background holes in each disocclusion and to define two spatial BG constraints that represent the lower and upper bounds of a background candidate. Holes are filled by comparing the similarities between the temporal background information and the spatial background constraints. In addition, a depth map-based ghost removal filter is described to solve the problem of the non-fit between a color image and the corresponding depth map of a virtual view. Finally, an exemplar-based inpainting is applied to fill in the remaining holes with a priority function that includes a new depth term. The experimental results demonstrated that the proposed method led to results that promised subjective and objective improvement over state-of-the-art methods.
- 11 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 21 (6) , 1345-1358
New Hole-Filling Method Using Extrapolated Spatio-Temporal Background Information for a Synthesized Free-View
This paper introduces a new hole-filling method using extrapolated spatio-temporal background information to obtain a synthesized free-view. New temporal background modeling is proposed, which incorporates stationary temporal information in the hole-filling process and preserves the temporal consistency of synthesized views. A background codebook is distinguished from a non-overlapped patch-based codebook, which contributes to extracting reliable temporal background information. Furthermore, a depth-map driven spatial local background estimation is also addressed to discriminate the background holes in each disocclusion and to define two spatial BG constraints that represent the lower and upper bounds of a background candidate. Holes are filled by comparing the similarities between the temporal background information and the spatial background constraints. In addition, a depth map-based ghost removal filter is described to solve the problem of the non-fit between a color image and the corresponding depth map of a virtual view. Finally, an exemplar-based inpainting is applied to fill in the remaining holes with a priority function that includes a new depth term. The experimental results demonstrated that the proposed method led to results that promised subjective and objective improvement over state-of-the-art methods.