Search Results for author: Sungjin Lee

Found 54 papers, 9 papers with code

Dialogue Response Generation via Contrastive Latent Representation Learning

no code implementations EMNLP (NLP4ConvAI) 2021 Shuyang Dai, Guoyin Wang, Sunghyun Park, Sungjin Lee

In this work, we aim to construct a robust sentence representation learning model, that is specifically designed for dialogue response generation, with Transformer-based encoder-decoder structure.

Contrastive Learning Decoder +3

Data-Efficient Alignment of Large Language Models with Human Feedback Through Natural Language

no code implementations24 Nov 2023 Di Jin, Shikib Mehri, Devamanyu Hazarika, Aishwarya Padmakumar, Sungjin Lee, Yang Liu, Mahdi Namazifar

Learning from human feedback is a prominent technique to align the output of large language models (LLMs) with human expectations.

Data Augmentation for Improving Tail-traffic Robustness in Skill-routing for Dialogue Systems

no code implementations7 Jun 2023 Ting-Wei Wu, Fatemeh Sheikholeslami, Mohammad Kachuee, Jaeyoung Do, Sungjin Lee

Large-scale conversational systems typically rely on a skill-routing component to route a user request to an appropriate skill and interpretation to serve the request.

Data Augmentation Decoder +1

Open World Classification with Adaptive Negative Samples

no code implementations9 Mar 2023 Ke Bai, Guoyin Wang, Jiwei Li, Sunghyun Park, Sungjin Lee, Puyang Xu, Ricardo Henao, Lawrence Carin

Open world classification is a task in natural language processing with key practical relevance and impact.

Classification

Cluster-Guided Label Generation in Extreme Multi-Label Classification

1 code implementation17 Feb 2023 Taehee Jung, Joo-Kyung Kim, Sungjin Lee, Dongyeop Kang

For extreme multi-label classification (XMC), existing classification-based models poorly perform for tail labels and often ignore the semantic relations among labels, like treating "Wikipedia" and "Wiki" as independent and separate labels.

Classification Extreme Multi-Label Classification

Grounding Counterfactual Explanation of Image Classifiers to Textual Concept Space

no code implementations CVPR 2023 Siwon Kim, Jinoh Oh, Sungjin Lee, Seunghak Yu, Jaeyoung Do, Tara Taghavi

In this paper, we propose counterfactual explanation with text-driven concepts (CounTEX), where the concepts are defined only from text by leveraging a pre-trained multi-modal joint embedding space without additional concept-annotated datasets.

counterfactual Counterfactual Explanation

Constrained Policy Optimization for Controlled Self-Learning in Conversational AI Systems

no code implementations17 Sep 2022 Mohammad Kachuee, Sungjin Lee

Based on the experimental results, we demonstrate that the proposed approach is capable of achieving the best balance between the policy value and constraint satisfaction rate.

Multi-Armed Bandits Self-Learning

Debiasing Neighbor Aggregation for Graph Neural Network in Recommender Systems

no code implementations18 Aug 2022 Minseok Kim, Jinoh Oh, Jaeyoung Do, Sungjin Lee

Graph neural networks (GNNs) have achieved remarkable success in recommender systems by representing users and items based on their historical interactions.

Graph Neural Network Recommendation Systems

Learning Personalized Representations using Graph Convolutional Network

no code implementations28 Jul 2022 Hongyu Shen, Jinoh Oh, Shuai Zhao, Guoyin Wang, Tara Taghavi, Sungjin Lee

Then we propose a graph convolutional network(GCN) based model, namely Personalized Dynamic Routing Feature Encoder(PDRFE), that generates personalized customer representations learned from the built graph.

Domain-Aware Contrastive Knowledge Transfer for Multi-domain Imbalanced Data

no code implementations WASSA (ACL) 2022 Zixuan Ke, Mohammad Kachuee, Sungjin Lee

In many real-world machine learning applications, samples belong to a set of domains e. g., for product reviews each review belongs to a product category.

Transfer Learning

DeepSketch: A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression

no code implementations17 Feb 2022 Jisung Park, Jeoggyun Kim, Yeseong Kim, Sungjin Lee, Onur Mutlu

Data reduction in storage systems is becoming increasingly important as an effective solution to minimize the management cost of a data center.

Management

Deciding Whether to Ask Clarifying Questions in Large-Scale Spoken Language Understanding

no code implementations25 Sep 2021 Joo-Kyung Kim, Guoyin Wang, Sungjin Lee, Young-Bum Kim

A large-scale conversational agent can suffer from understanding user utterances with various ambiguities such as ASR ambiguity, intent ambiguity, and hypothesis ambiguity.

Spoken Language Understanding

AUGNLG: Few-shot Natural Language Generation using Self-trained Data Augmentation

1 code implementation ACL 2021 Xinnuo Xu, Guoyin Wang, Young-Bum Kim, Sungjin Lee

Natural Language Generation (NLG) is a key component in a task-oriented dialogue system, which converts the structured meaning representation (MR) to the natural language.

Data Augmentation Retrieval +2

Learning Slice-Aware Representations with Mixture of Attentions

no code implementations Findings (ACL) 2021 Cheng Wang, Sungjin Lee, Sunghyun Park, Han Li, Young-Bum Kim, Ruhi Sarikaya

Real-world machine learning systems are achieving remarkable performance in terms of coarse-grained metrics like overall accuracy and F-1 score.

Natural Language Understanding

Handling Long-Tail Queries with Slice-Aware Conversational Systems

no code implementations26 Apr 2021 Cheng Wang, Sun Kim, Taiwoo Park, Sajal Choudhary, Sunghyun Park, Young-Bum Kim, Ruhi Sarikaya, Sungjin Lee

We have been witnessing the usefulness of conversational AI systems such as Siri and Alexa, directly impacting our daily lives.

Neural model robustness for skill routing in large-scale conversational AI systems: A design choice exploration

no code implementations4 Mar 2021 Han Li, Sunghyun Park, Aswarth Dara, Jinseok Nam, Sungjin Lee, Young-Bum Kim, Spyros Matsoukas, Ruhi Sarikaya

Ensuring model robustness or resilience in the skill routing component is an important problem since skills may dynamically change their subscription in the ontology after the skill routing model has been deployed to production.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +3

DEUS: A Data-driven Approach to Estimate User Satisfaction in Multi-turn Dialogues

no code implementations1 Mar 2021 Ziming Li, Dookun Park, Julia Kiseleva, Young-Bum Kim, Sungjin Lee

Digital assistants are experiencing rapid growth due to their ability to assist users with day-to-day tasks where most dialogues are happening multi-turn.

A scalable framework for learning from implicit user feedback to improve natural language understanding in large-scale conversational AI systems

no code implementations EMNLP 2021 Sunghyun Park, Han Li, Ameen Patel, Sidharth Mudgal, Sungjin Lee, Young-Bum Kim, Spyros Matsoukas, Ruhi Sarikaya

Natural Language Understanding (NLU) is an established component within a conversational AI or digital assistant system, and it is responsible for producing semantic understanding of a user request.

Natural Language Understanding

Guided Dialog Policy Learning without Adversarial Learning in the Loop

1 code implementation7 Apr 2020 Ziming Li, Sungjin Lee, Baolin Peng, Jinchao Li, Julia Kiseleva, Maarten de Rijke, Shahin Shayandeh, Jianfeng Gao

Reinforcement Learning (RL) methods have emerged as a popular choice for training an efficient and effective dialogue policy.

Reinforcement Learning (RL)

Structuring Latent Spaces for Stylized Response Generation

1 code implementation IJCNLP 2019 Xiang Gao, Yizhe Zhang, Sungjin Lee, Michel Galley, Chris Brockett, Jianfeng Gao, Bill Dolan

This structure allows the system to generate stylized relevant responses by sampling in the neighborhood of the conversation model prediction, and continuously control the style level.

Response Generation Style Transfer

Unsupervised Dialogue Spectrum Generation for Log Dialogue Ranking

no code implementations WS 2019 Xinnuo Xu, Yizhe Zhang, Lars Liden, Sungjin Lee

Although the data-driven approaches of some recent bot building platforms make it possible for a wide range of users to easily create dialogue systems, those platforms don{'}t offer tools for quickly identifying which log dialogues contain problems.

Contextual Out-of-Domain Utterance Handling With Counterfeit Data Augmentation

1 code implementation24 May 2019 Sungjin Lee, Igor Shalyminov

Neural dialog models often lack robustness to anomalous user input and produce inappropriate responses which leads to frustrating user experience.

Data Augmentation Out of Distribution (OOD) Detection

ConvLab: Multi-Domain End-to-End Dialog System Platform

2 code implementations ACL 2019 Sungjin Lee, Qi Zhu, Ryuichi Takanobu, Xiang Li, Yaoqin Zhang, Zheng Zhang, Jinchao Li, Baolin Peng, Xiujun Li, Minlie Huang, Jianfeng Gao

We present ConvLab, an open-source multi-domain end-to-end dialog system platform, that enables researchers to quickly set up experiments with reusable components and compare a large set of different approaches, ranging from conventional pipeline systems to end-to-end neural models, in common environments.

Consistent Dialogue Generation with Self-supervised Feature Learning

1 code implementation13 Mar 2019 Yizhe Zhang, Xiang Gao, Sungjin Lee, Chris Brockett, Michel Galley, Jianfeng Gao, Bill Dolan

Generating responses that are consistent with the dialogue context is one of the central challenges in building engaging conversational agents.

Dialogue Generation Response Generation

Jointly Optimizing Diversity and Relevance in Neural Response Generation

no code implementations NAACL 2019 Xiang Gao, Sungjin Lee, Yizhe Zhang, Chris Brockett, Michel Galley, Jianfeng Gao, Bill Dolan

In this paper, we propose a SpaceFusion model to jointly optimize diversity and relevance that essentially fuses the latent space of a sequence-to-sequence model and that of an autoencoder model by leveraging novel regularization terms.

Dialogue Generation Diversity +1

Improving Robustness of Neural Dialog Systems in a Data-Efficient Way with Turn Dropout

1 code implementation29 Nov 2018 Igor Shalyminov, Sungjin Lee

We present a new dataset for studying the robustness of dialog systems to OOD input, which is bAbI Dialog Task 6 augmented with OOD content in a controlled way.

Nudging Neural Conversational Model with Domain Knowledge

no code implementations15 Nov 2018 Sungjin Lee

Neural conversation models are attractive because one can train a model directly on dialog examples with minimal labeling.

Zero-Shot Adaptive Transfer for Conversational Language Understanding

no code implementations29 Aug 2018 Sungjin Lee, Rahul Jha

Conversational agents such as Alexa and Google Assistant constantly need to increase their language understanding capabilities by adding new domains.

Domain Adaptation

OneNet: Joint Domain, Intent, Slot Prediction for Spoken Language Understanding

no code implementations16 Jan 2018 Young-Bum Kim, Sungjin Lee, Karl Stratos

In practice, most spoken language understanding systems process user input in a pipelined manner; first domain is predicted, then intent and semantic slots are inferred according to the semantic frames of the predicted domain.

Spoken Language Understanding

Toward Continual Learning for Conversational Agents

no code implementations28 Dec 2017 Sungjin Lee

While end-to-end neural conversation models have led to promising advances in reducing hand-crafted features and errors induced by the traditional complex system architecture, they typically require an enormous amount of data due to the lack of modularity.

Continual Learning

Speaker-Sensitive Dual Memory Networks for Multi-Turn Slot Tagging

no code implementations29 Nov 2017 Young-Bum Kim, Sungjin Lee, Ruhi Sarikaya

In multi-turn dialogs, natural language understanding models can introduce obvious errors by being blind to contextual information.

Natural Language Understanding

Cannot find the paper you are looking for? You can Submit a new open access paper.