6 papers
Beyond On-Policy Exploration: Integrating External Policy Rollouts for Reinforcement Learning in Diffusion Language Models
Wonseok Lee, Jimyeong Kim, Jungmin Ko +1
Recent reinforcement learning methods for diffusion large language models (dLLMs) commonly rely on on-policy rollouts generated by the target dLLM itself. When successful on-policy…
Orthogonal Negative Guidance in Attention Feature Space for Text-to-Image Generation
Jungmin Ko, Jungwon Park, Jimyeong Kim +3
Text-to-image (T2I) models have become increasingly capable of generating high-quality images. Yet, enforcing the explicit absence of a specified object or attribute remains a fund…
When Confidence Misleads: Suffix Anchoring and Anchor-Proximity Confidence Modulation for Diffusion Language Models
Jungwon Park, Jimyeong Kim, Jungmin Ko +2
Diffusion language models generate text by iteratively selecting and denoising masked positions, making position selection a central inference-time decision. Most training-free met…
TLPO: Token-Level Policy Optimization for Mitigating Language Confusion in Large Language Models
Jinho Choo, JunSeung Lee, Jimyeong Kim +3
Large language models (LLMs) demonstrate strong multilingual capabilities, yet often fail to consistently generate responses in the intended language, exhibiting a phenomenon known…
Soft Head Selection for Injecting ICL-Derived Task Embeddings
Jungwon Park, Jimyeong Kim, Changin Choi +1
Large language models (LLMs) are commonly adapted to downstream tasks using parameter-efficient fine-tuning (PEFT) or in-context learning (ICL). Recently, ICL-driven embedding-base…
ReFlex: Text-Guided Editing of Real Images in Rectified Flow via Mid-Step Feature Extraction and Attention Adaptation
Jimyeong Kim, Jungwon Park, Yeji Song +2
Rectified Flow text-to-image models surpass diffusion models in image quality and text alignment, but adapting ReFlow for real-image editing remains challenging. We propose a new r…