VLM-to-DiT alignment in video editing models acts as a semantic bottleneck that degrades fine-grained structural semantics, demonstrated via a new diagnostic dataset and protocol on relation-based edits.
Kiwi-edit: Versatile video editing via instruction and reference guidance
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
abstract
Instruction-based video editing has witnessed rapid progress, yet current methods often struggle with precise visual control, as natural language is inherently limited in describing complex visual nuances. Although reference-guided editing offers a robust solution, its potential is currently bottlenecked by the scarcity of high-quality paired training data. To bridge this gap, we introduce a scalable data generation pipeline that transforms existing video editing pairs into high-fidelity training quadruplets, leveraging image generative models to create synthesized reference scaffolds. Using this pipeline, we construct RefVIE, a large-scale dataset tailored for instruction-reference-following tasks, and establish RefVIE-Bench for comprehensive evaluation. Furthermore, we propose a unified editing architecture, Kiwi-Edit, that synergizes learnable queries and latent visual features for reference semantic guidance. Our model achieves significant gains in instruction following and reference fidelity via a progressive multi-stage training curriculum. Extensive experiments demonstrate that our data and architecture establish a new state-of-the-art in controllable video editing. All datasets, models, and code is released at https://github.com/showlab/Kiwi-Edit.
citation-role summary
baseline 2
background 1
citation-polarity summary
years
2026 8verdicts
UNVERDICTED 8representative citing papers
Aurora introduces a VLM-based agent that converts raw user video edit requests into structured conditioning inputs for a unified diffusion transformer, improving performance on underspecified tasks via a new benchmark.
StreamingEffect enables real-time 720p human-centric video effect generation on one GPU via teacher-student distillation, keyframe control, and a new 130K video dataset.
Sparkle supplies a large-scale dataset and benchmark for instruction-driven video background replacement, enabling models that generate more natural and temporally consistent new scenes than earlier approaches.
Sound Sparks Motion is a test-time tuning approach that adjusts audio and text conditioning signals in multimodal video models using VLM feedback to produce specific motion edits while preserving content.
InsEdit adapts a video diffusion backbone for text-instruction video editing via Mutual Context Attention, achieving SOTA open-source results with O(100K) data while also supporting image editing.
ImVideoEdit learns video editing from 13K image pairs by decoupling spatial modifications from frozen temporal dynamics in pretrained models, matching larger video-trained systems in fidelity and consistency.
Mamoda2.5 is a 25B-parameter DiT-MoE unified AR-Diffusion model that reaches top video generation and editing benchmarks with 4-step inference up to 95.9x faster than baselines.
citing papers explorer
-
What Semantics Survive the Connector? Diagnosing VLM-to-DiT Alignment in Video Editing
VLM-to-DiT alignment in video editing models acts as a semantic bottleneck that degrades fine-grained structural semantics, demonstrated via a new diagnostic dataset and protocol on relation-based edits.
-
Aurora: Unified Video Editing with a Tool-Using Agent
Aurora introduces a VLM-based agent that converts raw user video edit requests into structured conditioning inputs for a unified diffusion transformer, improving performance on underspecified tasks via a new benchmark.
-
StreamingEffect: Real-Time Human-Centric Video Effect Generation
StreamingEffect enables real-time 720p human-centric video effect generation on one GPU via teacher-student distillation, keyframe control, and a new 130K video dataset.
-
Sparkle: Realizing Lively Instruction-Guided Video Background Replacement via Decoupled Guidance
Sparkle supplies a large-scale dataset and benchmark for instruction-driven video background replacement, enabling models that generate more natural and temporally consistent new scenes than earlier approaches.
-
Sound Sparks Motion: Audio and Text Tuning for Video Editing
Sound Sparks Motion is a test-time tuning approach that adjusts audio and text conditioning signals in multimodal video models using VLM feedback to produce specific motion edits while preserving content.
-
InsEdit: Towards Instruction-based Visual Editing via Data-Efficient Video Diffusion Models Adaptation
InsEdit adapts a video diffusion backbone for text-instruction video editing via Mutual Context Attention, achieving SOTA open-source results with O(100K) data while also supporting image editing.
-
ImVideoEdit: Image-learning Video Editing via 2D Spatial Difference Attention Blocks
ImVideoEdit learns video editing from 13K image pairs by decoupling spatial modifications from frozen temporal dynamics in pretrained models, matching larger video-trained systems in fidelity and consistency.
-
Mamoda2.5: Enhancing Unified Multimodal Model with DiT-MoE
Mamoda2.5 is a 25B-parameter DiT-MoE unified AR-Diffusion model that reaches top video generation and editing benchmarks with 4-step inference up to 95.9x faster than baselines.