Sungroh Yoon

MobileKGQA: On-Device KGQA System on Dynamic Mobile Environments

SSDi8: Accurate and Efficient 8-bit Quantization for State Space Duality

Negative-Guided Subject Fidelity Optimization for Zero-Shot Subject-Driven Generation

Hierarchical Periodic Stationarization for Non-stationary Time Series Forecasting

Adversarial Visual Contrastive Decoding for Mitigating Hallucinations in Large Vision-Language Models

Time Conditioned Foreseeing: Temporal Generative Pretraining for EHR foundation models

Enhancing the Representational Power of Spiking Neural Networks via Diversified Spike Patterns

TPI-VA: Third-Party Interruption-Aware Voice Assistant

Interpretable Next-token Prediction via the Generalized Induction Head

Interpretable Language Modeling via Induction-head Ngram Models

Visual Attention Never Fades: Selective Progressive Attention ReCalibration for Detailed Image Captioning in Multimodal Large Language Models

Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage

RePIC: Reinforced Post-Training for Personalizing Multi-Modal Language Models

Balancing Gradient Frequencies Facilitates Inductive Inference in Algorithmic Reasoning

Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection

Enhancing Factuality in Detailed Image Captioning with LLM-MLLM Collaboration