SCMAPR: Self-Correcting Multi-Agent Prompt Refinement for Complex-Scenario Text-to-Video Generation
This addresses the challenge of generating high-quality videos from complex text prompts for AI video generation applications, representing a novel method for a known bottleneck.
The paper tackles the problem of text-to-video generation struggling with complex scenarios due to ambiguous prompts by proposing SCMAPR, a multi-agent prompt refinement framework that improves text-video alignment and generation quality, achieving gains of up to 2.67% on VBench, 3.28 on EvalCrafter, and 0.028 on T2V-CompBench over state-of-the-art baselines.
Text-to-Video (T2V) generation has benefited from recent advances in diffusion models, yet current systems still struggle under complex scenarios, which are generally exacerbated by the ambiguity and underspecification of text prompts. In this work, we formulate complex-scenario prompt refinement as a stage-wise multi-agent refinement process and propose SCMAPR, i.e., a scenario-aware and Self-Correcting Multi-Agent Prompt Refinement framework for T2V prompting. SCMAPR coordinates specialized agents to (i) route each prompt to a taxonomy-grounded scenario for strategy selection, (ii) synthesize scenario-aware rewriting policies and perform policy-conditioned refinement, and (iii) conduct structured semantic verification that triggers conditional revision when violations are detected. To clarify what constitutes complex scenarios in T2V prompting, provide representative examples, and enable rigorous evaluation under such challenging conditions, we further introduce {T2V-Complexity}, which is a complex-scenario T2V benchmark consisting exclusively of complex-scenario prompts. Extensive experiments on 3 existing benchmarks and our T2V-Complexity benchmark demonstrate that SCMAPR consistently improves text-video alignment and overall generation quality under complex scenarios, achieving up to 2.67\% and 3.28 gains in average score on VBench and EvalCrafter, and up to 0.028 improvement on T2V-CompBench over 3 State-Of-The-Art baselines.