# CLEAR: Context-Aware Learning with End-to-End Mask-Free Inference for Adaptive Video Subtitle Removal

arXiv: https://arxiv.org/abs/2603.21901 · HTML: https://mlsanity.com/paper/2603.21901/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 29 references that are in the map.
- First posted on arXiv 23 Mar 2026.

## Builds on (most cited first, 25 of 25)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [Video Diffusion Models](https://mlsanity.com/paper/2204.03458/) (2022)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [Towards Accurate Generative Models of Video: A New Metric & Challenges](https://mlsanity.com/paper/1812.01717/) (2018)
- [Image Quality Assessment: Unifying Structure and Texture Similarity](https://mlsanity.com/paper/2004.07728/) (2020)
- [Learning Blind Video Temporal Consistency](https://mlsanity.com/paper/1808.00449/) (2018)
- [ProPainter: Improving Propagation and Transformer for Video Inpainting](https://mlsanity.com/paper/2309.03897/) (2023)
- [Deep Flow-Guided Video Inpainting](https://mlsanity.com/paper/1905.02884/) (2019)
- [VideoPainter: Any-length Video Inpainting and Editing with Plug-and-Play Context Control](https://mlsanity.com/paper/2503.05639/) (2025)
- [DiffuEraser: A Diffusion Model for Video Inpainting](https://mlsanity.com/paper/2501.10018/) (2025)
- [MiniMax-Remover: Taming Bad Noise Helps Video Object Removal](https://mlsanity.com/paper/2505.24873/) (2025)
- [Deep Learning-based Image and Video Inpainting: A Survey](https://mlsanity.com/paper/2401.03395/) (2024)
- [MTRNet: A Generic Scene Text Eraser](https://mlsanity.com/paper/1903.04092/) (2019)
- [Omni-Effects: Unified and Spatially-Controllable Visual Effects Generation](https://mlsanity.com/paper/2508.07981/) (2025)
- [MTRNet++: One-stage Mask-based Scene Text Eraser](https://mlsanity.com/paper/1912.07183/) (2019)
- [Choose What You Need: Disentangled Representation Learning for Scene Text Recognition, Removal and Editing](https://mlsanity.com/paper/2405.04377/) (2024)
- [MTV-Inpaint: Multi-Task Long Video Inpainting](https://mlsanity.com/paper/2503.11412/) (2025)
- [Understanding Attention Mechanism in Video Diffusion Models](https://mlsanity.com/paper/2504.12027/) (2025)
- [EraserDiT: Fast Video Inpainting with Diffusion Transformer Model](https://mlsanity.com/paper/2506.12853/) (2025)
- [DiffSTR: Controlled Diffusion Models for Scene Text Removal](https://mlsanity.com/paper/2410.21721/) (2024)
- [VIP: Video Inpainting Pipeline for Real World Human Removal](https://mlsanity.com/paper/2504.03041/) (2025)

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

- [SEDiT: Mask-Free Video Subtitle Erasure via One-step Diffusion Transformer](https://mlsanity.com/paper/2605.14894/) (2026)
- [MFVSR: A Multi-Style and Fine-Grained Benchmark for Video Subtitle Removal](https://mlsanity.com/paper/2610.18461/) (2026)

## Abstract (arXiv)

Video subtitle removal aims to distinguish text overlays from background content while preserving temporal coherence. Existing diffusion-based methods necessitate explicit mask sequences during both training and inference phases, which restricts their practical deployment. In this paper, we present CLEAR (Context-aware Learning for End-to-end Adaptive Video Subtitle Removal), a mask-free framework that achieves truly end-to-end inference through context-aware adaptive learning. Our two-stage design decouples prior extraction from generative refinement: Stage I learns disentangled subtitle representations via self-supervised orthogonality constraints on dual encoders, while Stage II employs LoRA-based adaptation with generation feedback for dynamic context adjustment. Notably, our method only requires 0.77% of the parameters of the base diffusion model for training. On Chinese subtitle benchmarks, CLEAR outperforms mask-dependent baselines by + 6.77dB PSNR and -74.7% VFID, while demonstrating superior zero-shot generalization across six languages (English, Korean, French, Japanese, Russian, German), a performance enabled by our generation-driven feedback mechanism that ensures robust subtitle removal without ground-truth masks during inference.