REVIEW 14 cited by
OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data
read the original abstract
Diffusion models have advanced image stylization significantly, yet two core challenges persist: (1) maintaining consistent stylization in complex scenes, particularly identity, composition, and fine details, and (2) preventing style degradation in image-to-image pipelines with style LoRAs. GPT-4o's exceptional stylization consistency highlights the performance gap between open-source methods and proprietary models. To bridge this gap, we propose \textbf{OmniConsistency}, a universal consistency plugin leveraging large-scale Diffusion Transformers (DiTs). OmniConsistency contributes: (1) an in-context consistency learning framework trained on aligned image pairs for robust generalization; (2) a two-stage progressive learning strategy decoupling style learning from consistency preservation to mitigate style degradation; and (3) a fully plug-and-play design compatible with arbitrary style LoRAs under the Flux framework. Extensive experiments show that OmniConsistency significantly enhances visual coherence and aesthetic quality, achieving performance comparable to commercial state-of-the-art model GPT-4o.
Forward citations
Cited by 14 Pith papers
-
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.
-
ST-BiBench: Benchmarking Multi-Stream Multimodal Coordination in Bimanual Embodied Tasks for MLLMs
ST-BiBench reveals a coordination paradox in which MLLMs show strong high-level strategic reasoning yet fail at fine-grained 16-dimensional bimanual action synthesis and multi-stream fusion.
-
Ghosts Beneath Textures: Texture-Relation Cues for Cross-Paradigm AI-Generated Image Detection
Semantics-irrelevant local-global texture relations, extracted after content suppression, serve as cross-paradigm forensic cues that let DTS-Det reach 99.6% accuracy on a new mixed-generation benchmark.
-
Style-CCL: Content-Preserving Style Transfer via Curriculum Continual Learning
Style-CCL uses curriculum continual learning on a million-scale synthetic dataset with a dual-branch SC-DiT to achieve state-of-the-art content-preserving style transfer.
-
PAI-Studio: Cinematic Video Background Replacement with Camera-Aware Motion
PAI-Studio reformulates cinematic background replacement as in-context conditional generation inside a Diffusion Transformer with bidirectional attention, trained on a new 30K film-sourced dataset, and reports better ...
-
CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation
A multi-teacher distillation framework that packs 50 effect LoRAs and fast sampling into a single adapter while aiming to avoid concept interference.
-
VISTA: Triplet-Supervised Video Style Transfer with Diffusion Transformers
VISTA introduces a new synthetic triplet dataset and diffusion-transformer framework with style adapter that jointly models style, content, and motion to achieve state-of-the-art video style transfer.
-
OmniHumanoid: Streaming Cross-Embodiment Video Generation with Paired-Free Adaptation
OmniHumanoid factorizes transferable motion learning from embodiment-specific adaptation to enable scalable cross-embodiment video generation without paired data for new humanoids.
-
ReMoT: Reinforcement Learning with Motion Contrast Triplets
Training a 4B vision-language model on rule-generated motion-contrast triplets with GRPO lifts spatio-temporal QA accuracy by about 17 points on the authors' own benchmark and by smaller margins on standard benchmarks.
-
SWEET: Sparse World Modeling with Image Editing for Embodied Task Execution
SWEET is a one-shot sparse visual planning framework that progressively generates manipulation keyframes via image editing conditioned on language and spatial guidance, then converts them to actions with a diffusion p...
-
AutoAWG: Adverse Weather Generation with Adaptive Multi-Controls for Automotive Videos
AutoAWG generates controllable adverse weather automotive videos via semantics-guided adaptive multi-control fusion and vanishing-point-anchored temporal synthesis from static images, reducing FID by 50% and FVD by 16...
-
WorldWander: Bridging Egocentric and Exocentric Worlds in Video Generation
A bidirectional egocentric-to-exocentric video translation framework trained with in-context attention on a new synthetic+real dataset, with evaluation flaws around reference leakage and missing direct baselines.
-
FeRA: Frequency-Energy Constrained Routing for Effective Diffusion Adaptation Fine-Tuning
A frequency-energy router that blends LoRA experts according to the latent's bandwise energy improves diffusion fine-tuning quality and style consistency across multiple backbones.
-
TeleStyle V2: Beyond Content-Preserving Style Transfer with Self-Distillation and Distribution-Matching-Distillation
TeleStyle V2 uses self-distillation from V1 plus DMD and a prompt enhancer to support RnR/RnS/SnR/SnS reference pairs while matching commercial models on style transfer and general editing.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.