1 citations · 2 across the 14 of their papers we have counts for
17 papers
Visual Grounding First, Multimodal In-context Learning Follows
Minhyuk Seo, Minjae Lee, Chaeeun Lee +4
LLMs have shown strong in-context learning (ICL) capability, but extending it to Multimodal LLMs (MLLMs) remains challenging. Prior multimodal ICL methods often rely on ICL-specifi…
Action Chunking Proximal Policy Optimization with Feedback Correction
Sanghyun Hahn, Jonghyun Choi
Action chunking provides temporal abstraction in reinforcement learning by selecting short action sequences instead of individual actions, but many existing approaches face two lim…
Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shifts
Taewook Kang, Taeheon Kim, Donghyun Shin +1
Vision-Language-Action (VLA) models often fail to perform the same learned tasks under environmental shifts, such as changes in camera pose and shifts to a different but similar ro…
Position: Modular Memory is the Key to Continual Learning Agents
Vaggelis Dorovatas, Malte Schwerin, Andrew D. Bagdanov +21
Foundation models have transformed machine learning through large-scale pretraining and increased test-time compute. Despite surpassing human performance in several domains, these…
Attention Misses Visual Risk: Risk-Adaptive Steering for Multimodal Safety Alignment
Jonghyun Park, Minhyuk Seo, Chaewon Yeo +1
Even modern AI models often remain vulnerable to multimodal queries in which harmful intent is embedded in images. A widely used approach for safety alignment is training with exte…
Multi-Level Knowledge Distillation and Dynamic Self-Supervised Learning for Continual Learning
Taeheon Kim, San Kim, Minhyuk Seo +3
Class-incremental with repetition (CIR), where previously trained classes repeatedly introduced in future tasks, is a more realistic scenario than the traditional class incremental…
TTA-DAME: Test-Time Adaptation with Domain Augmentation and Model Ensemble for Dynamic Driving Conditions
Dongjae Jeon, Taeheon Kim, Seongwon Cho +2
Test-time Adaptation (TTA) poses a challenge, requiring models to dynamically adapt and perform optimally on shifting target domains. This task is particularly emphasized in real-w…
Co-LoRA: Collaborative Model Personalization on Heterogeneous Multi-Modal Clients
Minhyuk Seo, Taeheon Kim, Hankook Lee +2
As AI becomes more personal, e.g., Agentic AI, there is an increasing need for personalizing models for various use cases. Personalized federated learning (PFL) enables each client…
OASIS: Online Sample Selection for Continual Visual Instruction Tuning
Minjae Lee, Minhyuk Seo, Tingyu Qu +2
In continual instruction tuning (CIT) scenarios, where new instruction tuning data continuously arrive in an online streaming manner, training delays from large-scale data signific…
Multi-Modal Grounded Planning and Efficient Replanning For Learning Embodied Agents with A Few Examples
Taewoong Kim, Byeonghwi Kim, Jonghyun Choi
Learning a perception and reasoning module for robotic assistants to plan steps to perform complex tasks based on natural language instructions often requires large free-form langu…
Large Language Models Still Exhibit Bias in Long Text
Wonje Jeung, Dongjae Jeon, Ashkan Yousefpour +1
Existing fairness benchmarks for large language models (LLMs) primarily focus on simple tasks, such as multiple-choice questions, overlooking biases that may arise in more complex…
Budgeted Online Continual Learning by Adaptive Layer Freezing and Frequency-based Sampling
Minhyuk Seo, Hyunseo Koh, Jonghyun Choi
The majority of online continual learning (CL) advocates single-epoch training and imposes restrictions on the size of replay memory. However, single-epoch training would incur a d…
ReALFRED: An Embodied Instruction Following Benchmark in Photo-Realistic Environments
Taewoong Kim, Cheolhong Min, Byeonghwi Kim +3
Simulated virtual environments have been widely used to learn robotic agents that perform daily household tasks. These environments encourage research progress by far, but often pr…
An Information Theoretic Evaluation Metric For Strong Unlearning
Dongjae Jeon, Wonje Jeung, Taeheon Kim +2
Machine unlearning (MU) aims to remove the influence of specific data from trained models, addressing privacy concerns and ensuring compliance with regulations such as the ``right…
Learning Equi-angular Representations for Online Continual Learning
Minhyuk Seo, Hyunseo Koh, Wonje Jeung +7
Online continual learning suffers from an underfitted solution due to insufficient training for prompt model update (e.g., single-epoch training). To address the challenge, we prop…
Online Continual Learning For Interactive Instruction Following Agents
Byeonghwi Kim, Minhyuk Seo, Jonghyun Choi
In learning an embodied agent executing daily tasks via language directives, the literature largely assumes that the agent learns all training data at the beginning. We argue that…
GenOL: Generating Diverse Examples for Name-only Online Learning
Minhyuk Seo, Seongwon Cho, Minjae Lee +4
Online learning methods often rely on supervised data. However, under data distribution shifts, such as in continual learning (CL), where continuously arriving online data streams…