5 citations · 9 across the 14 of their papers we have counts for
7 papers · 1 filter
LENS: In-Context Search via Latent Evidence Exploration over Dynamic Raw Documents
Xingjun Wang, Gongsheng Li, Qi Fan +3
LLM agents increasingly answer questions over dynamic raw-document collections, where files may change before preprocessing, and relevant evidence (spans, sections, pages, or table…
TreeAdapter: Hierarchical Taxonomy-Guided Adapter Composition for Fine-Grained Species Image Generation
Yuze Sun, Zhongjie Duan, Yingda Chen
Although general text-to-image models excel in open-domain generation, their performance degrades significantly in specialized downstream domains, particularly when generating imag…
Which Tokens Matter? Adaptive Token Selection for RLVR with the Relative Surprisal Index
Outongyi Lv, Yanzhao Zheng, Yuanwei Zhang +5
Reinforcement learning (RL) has become a powerful tool for propelling Large Language Models (LLMs) beyond imitation-based training towards more robust reasoning capabilities. Among…
Compressing Image Style Training into a Single Model Forward
Zhongjie Duan, Yingda Chen
Diffusion-based style transfer must balance inference efficiency with stylization fidelity. Adapter-based methods are efficient, but they inject style as an external condition and…
Diffusion Templates: A Unified Plugin Framework for Controllable Diffusion
Zhongjie Duan, Hong Zhang, Yingda Chen
Controllable diffusion methods have substantially expanded the practical utility of diffusion models, but they are typically developed as isolated, backbone-specific systems with i…
VIRAL: Visual In-Context Reasoning via Analogy in Diffusion Transformers
Zhiwen Li, Zhongjie Duan, Jinyan Ye +4
Replicating In-Context Learning (ICL) in computer vision remains challenging due to task heterogeneity. We propose \textbf{VIRAL}, a framework that elicits visual reasoning from a…
Spectral Evolution Search: Efficient Inference-Time Scaling for Reward-Aligned Image Generation
Jinyan Ye, Zhongjie Duan, Zhiwen Li +4
Inference-time scaling offers a versatile paradigm for aligning visual generative models with downstream objectives without parameter updates. However, existing approaches that opt…