# Self-Paced and Self-Corrective Masked Prediction for Movie Trailer Generation

arXiv: https://arxiv.org/abs/2512.04426 · HTML: https://mlsanity.com/paper/2512.04426/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 56 references that are in the map.
- First posted on arXiv 4 Dec 2025.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [ImageBind: One Embedding Space To Bind Them All](https://mlsanity.com/paper/2305.05665/) (2023)
- [MaskGIT: Masked Generative Image Transformer](https://mlsanity.com/paper/2202.04200/) (2022)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [VideoBERT: A Joint Model for Video and Language Representation Learning](https://mlsanity.com/paper/1904.01766/) (2019)
- [PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers](https://mlsanity.com/paper/2111.12710/) (2021)
- [TransNet V2: An effective deep network architecture for fast shot transition detection](https://mlsanity.com/paper/2008.04838/) (2020)
- [BEVT: BERT Pretraining of Video Transformers](https://mlsanity.com/paper/2112.01529/) (2021)
- [CLIP-It! Language-Guided Video Summarization](https://mlsanity.com/paper/2107.00650/) (2021)
- [Summarizing Videos with Attention](https://mlsanity.com/paper/1812.01969/) (2018)
- [Learning Trailer Moments in Full-Length Movies](https://mlsanity.com/paper/2008.08502/) (2020)
- [Finding the Right Moment: Human-Assisted Trailer Creation via Task Composition](https://mlsanity.com/paper/2111.08774/) (2021)
- [Towards Automated Movie Trailer Generation](https://mlsanity.com/paper/2404.03477/) (2024)
- [From Shots to Stories: LLM-Assisted Video Editing with Unified Language Representations](https://mlsanity.com/paper/2505.12237/) (2025)
- [Shot Sequence Ordering for Video Editing: Benchmarks, Metrics, and Cinematology-Inspired Computing Methods](https://mlsanity.com/paper/2503.17975/) (2025)
- [An Inverse Partial Optimal Transport Framework for Music-guided Movie Trailer Generation](https://mlsanity.com/paper/2407.19456/) (2024)
- [Video Timeline Modeling For News Story Understanding](https://mlsanity.com/paper/2309.13446/) (2023)

## Built on by (most cited first, 2 of 2)

- [GLANCE: A Global-Local Coordination Multi-Agent Framework for Music-Grounded Non-Linear Video Editing](https://mlsanity.com/paper/2604.05076/) (2026)
- [BEAT: Rhythm-Elastic Alignment for Agentic Music-guided Movie Trailer Generation](https://mlsanity.com/paper/2605.27067/) (2026)

## Abstract (arXiv)

As a challenging video editing task, movie trailer generation involves selecting and reorganizing movie shots to create engaging trailers. Currently, most existing automatic trailer generation methods employ a "selection-then-ranking" paradigm (i.e., first selecting key shots and then ranking them), which suffers from inevitable error propagation and limits the quality of the generated trailers. Beyond this paradigm, we propose a new self-paced and self-corrective masked prediction method called SSMP, which achieves state-of-the-art results in automatic trailer generation via bi-directional contextual modeling and progressive self-correction. In particular, SSMP trains a Transformer encoder that takes the movie shot sequences as prompts and generates corresponding trailer shot sequences accordingly. The model is trained via masked prediction, reconstructing each trailer shot sequence from its randomly masked counterpart. The mask ratio is self-paced, allowing the task difficulty to adapt to the model and thereby improving model performance. When generating a movie trailer, the model fills the shot positions with high confidence at each step and re-masks the remaining positions for the next prediction, forming a progressive self-correction mechanism that is analogous to how human editors work. Both quantitative results and user studies demonstrate the superiority of SSMP in comparison to existing automatic movie trailer generation methods. Demo is available at: https://github.com/Dixin-Lab/SSMP.