Copertina

REPA Works Until It Doesn't: Early-Stopped, Holistic Alignment Supercharges Diffusion Training

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Wang, Ziqiao, Zhao, Wangbo, Zhou, Yuhao, Li, Zekai, Liang, Zhiyuan, Shi, Mingjia, Zhao, Xuanlei, Zhou, Pengfei, Zhang, Kaipeng, Wang, Zhangyang, Wang, Kai, You, Yang
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computer Vision and Pattern Recognition Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Semantics at an Angle: When Cosine Similarity Works Until It Doesn't
di: You, Kisung
Pubblicazione: (2025)

Faster Vision Mamba is Rebuilt in Minutes via Merged Token Re-training
di: Shi, Mingjia, et al.
Pubblicazione: (2024)

Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights
di: Liang, Zhiyuan, et al.
Pubblicazione: (2025)

Active Learning Works! Until It Doesn't: Measuring the Effectiveness of Activity-Based Learning Exercises on Information Anxiety
di: Halpern, Rebecca
Pubblicazione: (2016)

Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism
di: Zhang, Haoxiang, et al.
Pubblicazione: (2026)

Enhance-A-Video: Better Generated Video for Free
di: Luo, Yang, et al.
Pubblicazione: (2025)

Consciousness Doesn't Do That
di: Matthias Michel
Pubblicazione: (2026)

Library Designs Revisited: What Works--What Doesn't.
di: Metz, T. John, et al.
Pubblicazione: (1987)

Order Doesn't Matter, But Reasoning Does: Training LLMs with Order-Centric Augmentation
di: He, Qianxi, et al.
Pubblicazione: (2025)

Reinforcement Learning for Reasoning in Small LLMs: What Works and What Doesn't
di: Dang, Quy-Anh, et al.
Pubblicazione: (2025)

The Usual Doesn't Work: Why We Need Problem-Based Learning
di: Spence, Larry
Pubblicazione: (2004)

In-Service and the School Library Media Specialist: What Works and What Doesn't.
di: Turner, Philip M.
Pubblicazione: (1988)

Chapter When It Doesn't Go to Plan
di: Kim, Seoyeon | https://orcid.org/0000-0002-2018-9520, et al.
Pubblicazione: (2026)

Why Johnny Can Read...But Doesn't
di: Landy, Sarah
Pubblicazione: (1977)

Teach AI What It Doesn't Know
di: Sean Du
Pubblicazione: (2026)

Prioritize Alignment in Dataset Distillation
di: Li, Zekai, et al.
Pubblicazione: (2024)

Dynamic Vision Mamba
di: Wu, Mengxuan, et al.
Pubblicazione: (2025)

Recurrent Diffusion for Large-Scale Parameter Generation
di: Wang, Kai, et al.
Pubblicazione: (2025)

When the Conversation Doesn't Go Your Way
Pubblicazione: (2024)

Sometimes the Roof Doesn't Just Leak, It Caves In!
di: Curran, Charles, et al.
Pubblicazione: (1999)

RAPID^3: Tri-Level Reinforced Acceleration Policies for Diffusion Transformer
di: Zhao, Wangbo, et al.
Pubblicazione: (2025)

AG-REPA: Causal Layer Selection for Representation Alignment in Audio Flow Matching
di: Zhang, Pengfei, et al.
Pubblicazione: (2026)

SFT Doesn't Always Hurt General Capabilities: Revisiting Domain-Specific Fine-Tuning in LLMs
di: Lin, Jiacheng, et al.
Pubblicazione: (2025)

Doesn't Everyone Have Rights to a Learner's Permit?
di: Gehrig, Jody, et al.
Pubblicazione: (2009)

A Teacher's Influence Doesn't Apply Only to Students
di: Coatney, Sharon
Pubblicazione: (2005)

Staff Development in Libraries: Why It Frequently Doesn't Take.
di: Shaughnessy, Thomas W.
Pubblicazione: (1988)

Supercharging Graph Transformers with Advective Diffusion
di: Wu, Qitian, et al.
Pubblicazione: (2023)

Infinite Width Models That Work: Why Feature Learning Doesn't Matter as Much as You Think
di: Sernau, Luke
Pubblicazione: (2024)

Real-Time Video Generation with Pyramid Attention Broadcast
di: Zhao, Xuanlei, et al.
Pubblicazione: (2024)

When Structure Doesn't Help: LLMs Do Not Read Text-Attributed Graphs as Effectively as We Expected
di: Xu, Haotian, et al.
Pubblicazione: (2025)

The Clock Doesn't Close: Euler-Mascheroni as Torus Non-Closure
di: David Jan Lowder, et al.
Pubblicazione: (2026)

Against Parfitian Aggregation: Why Reductionism Doesn't Save Utilitarianism
di: Biagi, Tommaso
Pubblicazione: (2026)

Library Learning Doesn't: The Curious Case of the Single-Use "Library"
di: Berlot-Attwell, Ian, et al.
Pubblicazione: (2024)

Korea's Solidarity with the Global South (to Which It Didn't and Doesn't Belong)*
di: Suweon Kim
Pubblicazione: (2025)

Something There Is That Doesn't Love a Computer (Nor Hate It Either).
di: Friedman, Fred T.
Pubblicazione: (1984)

Why Prompt Optimization Works, and Why It Sometimes Doesn't: A Causal-Inspired Edit-Level Analysis
di: Gong, Shuzhi, et al.
Pubblicazione: (2026)

"The Law Doesn't Work Like a Computer": Exploring Software Licensing Issues Faced by Legal Practitioners
di: Wintersgill, Nathan, et al.
Pubblicazione: (2024)

Camera Height Doesn't Change: Unsupervised Training for Metric Monocular Road-Scene Depth Estimation
di: Kinoshita, Genki, et al.
Pubblicazione: (2023)

Neural-Driven Image Editing
di: Zhou, Pengfei, et al.
Pubblicazione: (2025)

Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields
di: Zhou, Shijie, et al.
Pubblicazione: (2023)