10 papers
Do LLMs Learn from Rewards in Context? : Rethinking the role of reward in In-Context RL
Minchan Kwon, Seunghee Koh, Sunghyun Baek +2
LLM agents increasingly improve at inference time by accumulating experience in context rather than by updating parameters. This process is often described as in-context reinforcem…
VGGT-Bridge: Beyond Sequential Pose Graphs via Coarse-Stride Skip Edges
Sungjae Choi, Hanna Bae, Sunghyun Baek +1
Feed-forward visual geometry transformers such as VGGT reconstruct dense 3D structure from images in a single forward pass, simplifying multi-view 3D reconstruction. However, their…
Info3R: Information-Adaptive Test-Time Training for 3D Reconstruction
Sunghyun Baek, Hanna Bae, Minchan Kwon +1
Transformer-based models have recently achieved strong performance on 3D reconstruction from images, and recent works extend them to process video streams in an online manner for r…
Direct Diversity Optimization for Diverse Successful Trajectories in Preference Post-Training
Junwon Ko, Dong-Jae Lee, Minchan Kwon +2
LLM agents for sequential decision tasks are often post-trained with trajectory-level outcome labels, but such labels provide little supervision for preserving multiple successful…
</think> Doesn't Stop Reasoning: Analysis of Spurious CoT Termination
Seunghee Koh, Sungjae Choi, Minchan Kwon +2
Chain-of-thought (CoT) reasoning improves large reasoning models (LRMs) on complex tasks but often produces long, redundant traces. Recent training-free early-exit methods shorten…
Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance
Minchan Kwon, Sunghyun Baek, Minseo Kim +3
Large Language Model (LLM) Red-Teaming, which proactively identifies vulnerabilities of LLMs, is an essential process for ensuring safety. Finding effective and diverse attacks in…
Forget What Matters, Keep the Rest: Selective Unlearning of Informative Tokens
Seunghee Koh, Sunghyun Baek, Youngdong Kim +1
Unlearning in large language models (LLMs) has emerged as a promising safeguard against adversarial behaviors. When the forgetting loss is applied uniformly without considering tok…
IMSE: Intrinsic Mixture of Spectral Experts Fine-tuning for Test-Time Adaptation
Sunghyun Baek, Jaemyung Yu, Seunghee Koh +3
Test-time adaptation (TTA) has been widely explored to prevent performance degradation when test data differ from the training distribution. However, fully leveraging the rich repr…
DAM: Domain-Aware Module for Multi-Domain Dataset Condensation
Jaehyun Choi, Gyojin Han, Dong-Jae Lee +2
Dataset Condensation (DC) has emerged as a promising solution to mitigate the computational and storage burdens associated with training deep learning models. However, existing DC…
AI-KD: Adversarial learning and Implicit regularization for self-Knowledge Distillation
Hyungmin Kim, Sungho Suh, Sunghyun Baek +4
We present a novel adversarial penalized self-knowledge distillation method, named adversarial learning and implicit regularization for self-knowledge distillation (AI-KD), which r…