Donghyun Kim

27

Papers

52

Total Citations

Papers (27)

Visual Delta Generator with Large Multi-modal Models for Semi-supervised Composed Image Retrieval

What How and When Should Object Detectors Update in Continually Changing Test Domains?

Adaptive Self-training Framework for Fine-grained Scene Graph Generation

RA-SGG: Retrieval-Augmented Scene Graph Generation Framework via Multi-Prototype Learning

Difference Inversion: Interpolate and Isolate the Difference with Token Consistency for Image Analogy Generation

LLM4SGG: Large Language Models for Weakly Supervised Scene Graph Generation

Excitation Backprop for RNNs

CODA-Prompt: COntinual Decomposed Attention-Based Prompting for Rehearsal-Free Continual Learning

ConStruct-VL: Data-Free Continual Structured VL Concepts Learning

CDS: Cross-Domain Self-Supervised Pre-Training

Learning Cross-Modal Contrastive Features for Video Domain Adaptation

Tune It the Right Way: Unsupervised Validation of Domain Adaptation via Soft Neighborhood Density

Going Beyond Nouns With Vision & Language Models Using Synthetic Data

CDAC: Cross-domain Attention Consistency in Transformer for Domain Adaptive Semantic Segmentation

Learning to Scale Multilingual Representations for Vision-Language Tasks

A Unified Framework for Domain Adaptive Pose Estimation

A Broad Study of Pre-training for Domain Generalization and Adaptation

Semi-Supervised Domain Adaptation via Minimax Entropy

MA-CIR: A Multimodal Arithmetic Benchmark for Composed Image Retrieval

ZIUM: Zero-Shot Intent-Aware Adversarial Attack on Unlearned Models

CaptionSmiths: Flexibly Controlling Language Pattern in Image Captioning

PLATYPUS: Progressive Local Surface Estimator for Arbitrary-Scale Point Cloud Upsampling

Universal Domain Adaptation through Self Supervision

OpenMatch: Open-Set Semi-supervised Learning with Open-set Consistency Regularization

Towards Efficient Image Compression Without Autoregressive Models

Learning Human Action Recognition Representations Without Real Humans

Dense and Aligned Captions (DAC) Promote Compositional Reasoning in VL Models