11 papers
When Should Enterprise Policy Decisions Be Learned, and When Should They Be Reasoned?
Hyojung Han, Ryangkyung Kang, Youngeun Kim +1
An enterprise agent that answers under a written policy must choose an action before it says anything: answer, ask for a missing fact, ask for confirmation, call one or several too…
ALoDLM: Adaptively Looped Diffusion Language Models
Liancheng Fang, Zhuowei Li, Youngeun Kim +10
Diffusion language models (DLMs) enable fast generation by predicting multiple tokens in parallel, but their practical adoption remains limited by a persistent quality gap relative…
Deferred Audio Pruning with Local Audio-Visual Dynamics for Omni-LLMs
Kyeongyoon Lee, Hongyeob Kim, Youngeun Kim +1
Omni-modal LLMs jointly process audio, video, and text, but long multimodal sequences incur substantial prefill and KV-cache costs. Existing omni-modal compression methods primaril…
SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agents
Ryangkyung Kang, Hongcheol Cho, Youngeun Kim
As LLM agents are increasingly deployed with large libraries of reusable skills, selecting the right skill for a user request has become a critical systems challenge. In small libr…
Real-Time Visual Attribution Streaming in Thinking Model
Seil Kang, Woojung Han, Junhyeok Kim +3
We present an amortized framework for real-time visual attribution streaming in multimodal thinking models. When these models generate code from a screenshot or solve math problems…
VisRef: Visual Refocusing while Thinking Improves Test-Time Scaling in Multi-Modal Large Reasoning Models
Soumya Suvra Ghosal, Youngeun Kim, Zhuowei Li +6
Advances in large reasoning models have shown strong performance on complex reasoning tasks by scaling test-time compute through extended reasoning. However, recent studies observe…
MC-GRPO: Median-Centered Group Relative Policy Optimization for Small-Rollout Reinforcement Learning
Youngeun Kim
Group-relative policy optimization methods train language models by generating multiple rollouts per prompt and normalizing rewards with a shared mean reward baseline. In resource-…
ZOO-Prune: Training-Free Token Pruning via Zeroth-Order Gradient Estimation in Vision-Language Models
Youngeun Kim, Youjia Zhang, Huiling Liu +3
Large Vision-Language Models (VLMs) enable strong multimodal reasoning but incur heavy inference costs from redundant visual tokens. Token pruning alleviates this issue, yet existi…
Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment
Youjia Zhang, Youngeun Kim, Young-Geun Choi +3
Test-time adaptation (TTA) enhances the zero-shot robustness under distribution shifts by leveraging unlabeled test data during inference. Despite notable advances, several challen…
Task Vector Quantization for Memory-Efficient Model Merging
Youngeun Kim, Seunghwan Lee, Aecheon Jung +2
Model merging enables efficient multi-task models by combining task-specific fine-tuned checkpoints. However, storing multiple task-specific checkpoints requires significant memory…
Open-World Dynamic Prompt and Continual Visual Representation Learning
Youngeun Kim, Jun Fang, Qin Zhang +7
The open world is inherently dynamic, characterized by ever-evolving concepts and distributions. Continual learning (CL) in this dynamic open-world environment presents a significa…