4 papers
VARI: Robust Text-to-Video Generation via Group-Structured Intent Alignment
Mingxuan Zhu, Hohei Chan, Shiyuan Liu +2
Modern text-to-video (T2V) models generate visually compelling videos, yet meaning-preserving changes in prompt wording can substantially alter semantic realization and overall qua…
FlowCred: Hierarchical Credit Assignment and Functional Coordination for Joint Text-to-Video Post-Training
Mingxuan Zhu, Yanwei Huang, Shiyuan Liu +4
Prompt optimization and generator post-training are two complementary ways to improve text-to-video (T2V) generation, and recent systems increasingly train both modules jointly. Ho…
LigBench: A Unified and Human-Aligned Benchmark for LLM-based Research Idea Generation
Chenrun Wang, Mingxuan Zhu, Tiancheng Huang +6
With the rapid advancement of large language models (LLMs), research idea generation has attracted increasing attention. Existing approaches enable LLMs to retrieve relevant litera…
IEA: Amateur-Friendly Conversational Image Editing Agent via Three Stages of Multitask Alignment
Zichen Zhu, Yuheng Sun, Mingxuan Zhu +11
Current image editing software often hinges on fixed filters or expert tuning, leaving a gap between amateur users' intent and outcomes. Creations by generative models may contain…