2 citations · 7 across the 21 of their papers we have counts for
21 papers
From Awareness to Adherence: Bridging the Context Gap in Spoken Dialogue Systems via Context-Aware Decoding
Che Hyun Lee, Heeseung Kim, Sungroh Yoon
Despite the success of end-to-end (E2E) spoken dialogue systems, maintaining strict context adherence in multi-round conversations remains a challenge. While prior works attribute…
R2R2: Robust Representation for Intensive Experience Reuse via Redundancy Reduction in Self-Predictive Learning
Sanghyeob Song, Donghyeok Lee, Jinsik Kim +1
For reinforcement learning in data-scarce domains like real-world robotics, intensive data reuse enhances efficiency but induces overfitting. While prior works focus on critic bias…
Diagnosing and Correcting Concept Omission in Multimodal Diffusion Transformers
Kanghyun Baek, Jaihyun Lew, Chaehun Shin +2
Multimodal Diffusion Transformers (MM-DiTs) have achieved remarkable progress in text-to-image generation, yet they frequently suffer from concept omission, where specified objects…
Omni-Persona: Systematic Benchmarking and Improving Omnimodal Personalization
Yeongtak Oh, Dongwook Lee, Sangkwon Park +2
While multimodal large language models have advanced across text, image, and audio, personalization research has remained primarily vision-language, with unified omnimodal benchmar…
Knowledge Integration Decay in Search-Augmented Reasoning of Large Language Models
Sangwon Yu, Ik-hwan Kim, Donghun Kang +6
Modern Large Language Models (LLMs) have demonstrated remarkable capabilities in complex tasks by employing search-augmented reasoning to incorporate external knowledge into long c…
Distilling Metacognitive Behaviors through Agentic Editing
Ik-hwan Kim, Hyeongrok Han, Mingi Jung +5
Distilling long reasoning traces from a teacher model into a smaller student also passes on the teacher's reasoning habits. Prior work and our annotation of teacher traces show tha…
A Multifaceted Analysis of Negative Bias in Large Language Models through the Lens of Parametric Knowledge
Jongyoon Song, Sangwon Yu, Sungroh Yoon
Negative bias refers to the tendency of large language models (LLMs) to excessively generate negative responses in binary decision tasks (e.g., yes-no question answering). Previous…
Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage
Saehyung Lee, Seunghyun Yoon, Trung Bui +2
Multimodal large language models (MLLMs) excel at generating highly detailed captions but often produce hallucinations. Our analysis reveals that existing hallucination detection m…
Unleashing Multi-Hop Reasoning Potential in Large Language Models through Repetition of Misordered Context
Sangwon Yu, Ik-hwan Kim, Jongyoon Song +3
Multi-hop reasoning, which requires multi-step reasoning based on the supporting documents within a given context, remains challenging for large language models (LLMs). LLMs often…
Correcting Negative Bias in Large Language Models through Negative Attention Score Alignment
Sangwon Yu, Jongyoon Song, Bongkyu Hwang +7
A binary decision task, like yes-no questions or answer verification, reflects a significant real-world scenario such as where users look for confirmation about the correctness of…
Interactive Text-to-Image Retrieval with Large Language Models: A Plug-and-Play Approach
Saehyung Lee, Sangwon Yu, Junsung Park +2
In this paper, we primarily address the issue of dialogue-form context query within the interactive text-to-image retrieval task. Our methodology, PlugIR, actively utilizes the gen…
Large Language Models are Skeptics: False Negative Problem of Input-conflicting Hallucination
Jongyoon Song, Sangwon Yu, Sungroh Yoon
In this paper, we identify a new category of bias that induces input-conflicting hallucinations, where large language models (LLMs) generate responses inconsistent with the content…
Improving Diffusion-Based Generative Models via Approximated Optimal Transport
Daegyu Kim, Jooyoung Choi, Chaehun Shin +2
We introduce the Approximated Optimal Transport (AOT) technique, a novel training scheme for diffusion-based generative models. Our approach aims to approximate and integrate optim…
Entity-level Factual Adaptiveness of Fine-tuning based Abstractive Summarization Models
Jongyoon Song, Nohil Park, Bongkyu Hwang +4
Abstractive summarization models often generate factually inconsistent content particularly when the parametric knowledge of the model conflicts with the knowledge in the input doc…
GrounDial: Human-norm Grounded Safe Dialog Response Generation
Siwon Kim, Shuyang Dai, Mohammad Kachuee +3
Current conversational AI systems based on large language models (LLMs) are known to generate unsafe responses, agreeing to offensive user input or including toxic content. Previou…
STAG: Structural Test-time Alignment of Gradients for Online Adaptation
Juhyeon Shin, Yujin Oh, Jonghyun Lee +5
Test-Time Adaptation (TTA) adapts pre-trained models using only unlabeled test streams, requiring real-time inference and update without access to source data. We propose Structura…
Paralinguistics-Aware Speech-Empowered Large Language Models for Natural Conversation
Heeseung Kim, Soonshin Seo, Kyeongseok Jeong +9
Recent work shows promising results in expanding the capabilities of large language models (LLM) to directly understand and synthesize speech. However, an LLM-based strategy for mo…
DAFA: Distance-Aware Fair Adversarial Training
Hyungyu Lee, Saehyung Lee, Hyemi Jang +3
The disparity in accuracy between classes in standard training is amplified during adversarial training, a phenomenon termed the robust fairness problem. Existing methodologies aim…
On mitigating stability-plasticity dilemma in CLIP-guided image morphing via geodesic distillation loss
Yeongtak Oh, Saehyung Lee, Uiwon Hwang +1
Large-scale language-vision pre-training models, such as CLIP, have achieved remarkable text-guided image morphing results by leveraging several unconditional generative models. Ho…
On the Analysis of Cross-Lingual Prompt Tuning for Decoder-based Multilingual Model
Nohil Park, Joonsuk Park, Kang Min Yoo +1
An exciting advancement in the field of multilingual models is the emergence of autoregressive models with zero- and few-shot capabilities, a phenomenon widely reported in large-sc…
Controlled Text Generation for Black-box Language Models via Score-based Progressive Editor
Sangwon Yu, Changmin Lee, Hojin Lee +1
Controlled text generation is very important for the practical use of language models because it ensures that the produced text includes only the desired attributes from a specific…