MCSC-Bench: Multimodal Context-to-Script Creation for Realistic Video Production
For researchers in video production and multimodal AI, this benchmark fills a gap by evaluating the complete script creation process, but it is an incremental contribution as it extends existing sub-task benchmarks.
The paper introduces MCSC-Bench, a benchmark for multimodal context-to-script creation that evaluates the full video production workflow from noisy inputs to structured scripts. Experiments show that current multimodal LLMs struggle with this task, but models trained on MCSC-Bench achieve SOTA, with an 8B model outperforming Gemini-2.5-Pro.
Real-world video creation often involves a complex reasoning workflow of selecting relevant shots from noisy materials, planning missing shots for narrative completeness, and organizing them into coherent storylines. However, existing benchmarks focus on isolated sub-tasks and lack support for evaluating this full process. To address this gap, we propose Multimodal Context-to-Script Creation (MCSC), a new task that transforms noisy multimodal inputs and user instructions into structured, executable video scripts. We further introduce MCSC-Bench, the first large-scale MCSC dataset, comprising 11K+ well-annotated videos. Each sample includes: (1) redundant multimodal materials and user instructions; (2) a coherent, production-ready script containing material-based shots, newly planned shots (with shooting instructions), and shot-aligned voiceovers. MCSC-Bench supports comprehensive evaluation across material selection, narrative planning, and conditioned script generation, and includes both in-domain and out-of-domain test sets. Experiments show that current multimodal LLMs struggle with structure-aware reasoning under long contexts, highlighting the challenges posed by our benchmark. Models trained on MCSC-Bench achieve SOTA performance, with an 8B model surpassing Gemini-2.5-Pro, and generalize to out-of-domain scenarios. Downstream video generation guided by the generated scripts further validates the practical value of MCSC. Datasets will be public soon.