Sung Ju Hwang

56

Papers

240

Total Citations

Papers (56)

AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML

Expanding Object Detector's Horizon: Incremental Learning Framework for Object Detection in Videos

Learning Diverse Attacks on Large Language Models for Robust Red-Teaming and Safety Tuning

ECLIPSE: Efficient Continual Learning in Panoptic Segmentation with Visual Prompt Tuning

DiffusionNAG: Predictor-guided Neural Architecture Generation with Diffusion Models

VideoICL: Confidence-based Iterative In-context Learning for Out-of-Distribution Video Understanding

Distilling LLM Agent into Small Models with Retrieval and Code Tools

Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model

A Training-Free Sub-quadratic Cost Transformer Model Serving Framework with Hierarchically Pruned Attention

System Prompt Optimization with Meta-Learning

Training Free Exponential Context Extension via Cascading KV Cache

Progressive Fourier Neural Representation for Sequential Video Compilation

Delta Attention: Fast and Accurate Sparse Attention Inference by Delta Correction

Text-Conditioned Sampling Framework for Text-to-Image Generation with Masked Generative Models

Silent Branding Attack: Trigger-free Data Poisoning Attack on Text-to-Image Diffusion Models

Bayesian Neural Scaling Law Extrapolation with Prior-Data Fitted Networks

Graph Generation with Diffusion Mixture

Drug Discovery with Dynamic Goal-aware Fragments

One Prompt is not Enough: Automated Construction of a Mixture-of-Expert Prompts

Generative Modeling on Manifolds Through Mixture of Riemannian Diffusion Processes

EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens

STELLA: Continual Audio-Video Pre-training with SpatioTemporal Localized Alignment

BECoTTA: Input-dependent Online Blending of Experts for Continual Test-time Adaptation

Learning to Quantize Deep Networks by Optimizing Quantization Intervals With Task Loss

MPViT: Multi-Path Vision Transformer for Dense Prediction

The Devil Is in the Points: Weakly Semi-Supervised Instance Segmentation via Point-Guided Mask Representation

Cluster-Promoting Quantization With Bit-Drop for Minimizing Network Quantization Loss

Combined Group and Exclusive Sparsity for Deep Neural Networks

Deep Asymmetric Multi-task Feature Learning

Learning What and Where to Transfer

Uncertainty-Aware Attention for Reliable Interpretation and Prediction

DropMax: Adaptive Variational Softmax

Joint Active Feature Acquisition and Classification with Variable-Size Set Encoding

Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Prediction

Adversarial Self-Supervised Contrastive Learning

Attribution Preservation in Network Compression for Reliable Network Interpretation

Bootstrapping neural processes

Neural Complexity Measures

MetaPerturb: Transferable Regularizer for Heterogeneous Tasks and Architectures

Distribution Aligning Refinery of Pseudo-label for Imbalanced Semi-supervised Learning

Few-shot Visual Reasoning with Meta-Analogical Contrastive Learning

Time-Reversal Symmetric ODE Network

Edge Representation Learning with Hypergraphs

Hit and Lead Discovery with Explorative RL and Fragment-based Molecule Generation

Task-Adaptive Neural Network Search with Meta-Contrastive Learning

Mini-Batch Consistent Slot Set Encoder for Scalable Set Encoding

Hardware-adaptive Efficient Latency Prediction for NAS via Meta-Learning

Set-based Meta-Interpolation for Few-Task Meta-Learning

Graph Self-supervised Learning with Accurate Discrepancy Learning

Learning to Generate Inversion-Resistant Model Explanations

Factorized-FL: Personalized Federated Learning with Parameter Factorization & Similarity Matching

STXD: Structural and Temporal Cross-Modal Distillation for Multi-View 3D Object Detection

Generalizable Lightweight Proxy for Robust NAS against Diverse Perturbations

Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks

Effective Targeted Attacks for Adversarial Self-Supervised Learning

SplitNet: Learning to Semantically Split Deep Networks for Parameter Reduction and Model Parallelization