613 citations · 1.2k across the 67 of their papers we have counts for
11 papers · 1 filter
SyMerge: From Non-Interference to Synergistic Merging via Single-Layer Adaptation
Aecheon Jung, Seunghwan Lee, Dongyoon Han +1
Model merging combines independently trained models into a single multi-task model. However, most existing approaches focus primarily on avoiding task interference. We argue that i…
MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation
Minhyun Lee, Seungho Lee, Song Park +3
Referring Image Segmentation (RIS) is an advanced vision-language task that involves identifying and segmenting objects within an image as described by free-form text descriptions.…
NegMerge: Sign-Consensual Weight Merging for Machine Unlearning
Hyo Seo Kim, Dongyoon Han, Junsuk Choe
Machine unlearning aims to selectively remove specific knowledge from a trained model. Existing approaches, such as Task Arithmetic, fine-tune the model on the forget set to create…
DaWin: Training-free Dynamic Weight Interpolation for Robust Adaptation
Changdae Oh, Yixuan Li, Kyungwoo Song +2
Adapting a pre-trained foundation model on downstream tasks should ensure robustness against distribution shifts without the need to retrain the whole model. Although existing weig…
Token-Supervised Value Models for Enhancing Mathematical Problem-Solving Capabilities of Large Language Models
Jung Hyun Lee, June Yong Yang, Byeongho Heo +4
With the rapid advancement of test-time compute search strategies to improve the mathematical problem-solving capabilities of large language models (LLMs), the need for building ro…
HYPE: Hyperbolic Entailment Filtering for Underspecified Images and Texts
Wonjae Kim, Sanghyuk Chun, Taekyung Kim +2
In an era where the volume of data drives the effectiveness of self-supervised learning, the specificity and clarity of data semantics play a crucial role in model training. Addres…
Leveraging Temporal Contextualization for Video Action Recognition
Minji Kim, Dongyoon Han, Taekyung Kim +1
We propose a novel framework for video understanding, called Temporally Contextualized CLIP (TC-CLIP), which leverages essential temporal information through global interactions in…
DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs
Donghyun Kim, Byeongho Heo, Dongyoon Han
This paper revives Densely Connected Convolutional Networks (DenseNets) and reveals the underrated effectiveness over predominant ResNet-style architectures. We believe DenseNets'…
Model Stock: All we need is just a few fine-tuned models
Dong-Hwan Jang, Sangdoo Yun, Dongyoon Han
This paper introduces an efficient fine-tuning method for large pre-trained models, offering strong in-distribution (ID) and out-of-distribution (OOD) performance. Breaking away fr…
Rotary Position Embedding for Vision Transformer
Byeongho Heo, Song Park, Dongyoon Han +1
Rotary Position Embedding (RoPE) performs remarkably on language models, especially for length extrapolation of Transformers. However, the impacts of RoPE on computer vision domain…
Morphing Tokens Draw Strong Masked Image Models
Taekyung Kim, Byeongho Heo, Dongyoon Han
Masked image modeling (MIM) has emerged as a promising approach for pre-training Vision Transformers (ViTs). MIMs predict masked tokens token-wise to recover target signals that ar…