#Training-Free

9 статей
arXiv Computer Vision
arXiv Computer Vision

AD-Relight: Training-Free Banner Relighting через Illution Translate with Diffication Pors

arXiv:2604244707v2 Annuales Type: заменить резюме: Недавний всплеск потребления контента через ретрансляционные услуги привел к росту спроса на персональное содержание. Личные объявления (ад) играют р...

arXiv Computer Vision
4
Читать
arXiv Machine Learning
arXiv Machine Learning

ExFold: Unified Expert Folding for Training-Free MoE Prefill-Decode Acceleration

arXiv:2608.24938v1 Анонс Тип: новый Абстракт: Модели Смеси Экспертов (MoE) масштабируют емкость для высокого качества, сохраняя при этом вычисления для каждого токена ограниченными через скудную актив...

arXiv Machine Learning
3
Читать
arXiv Computer Vision
arXiv Computer Vision

SAM2Dual: Training-Free, Dual Memory for Long-Term Video Object Segmentation

arXiv:2608.18640v1 Announce Type: new Abstract: Long-term video object segmentation (VOS) remains challenging due to error accumulation under extended occlusions, re-appearance, and scene changes. Alt...

arXiv Computer Vision
4
Читать
arXiv Computer Vision
arXiv Computer Vision

Partition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and World Models

arXiv:2608.18484v1 Announce Type: new Abstract: Training-free block-sparse attention can accelerate video transformers, but row-wise attention concentration does not by itself specify an executable sp...

arXiv Computer Vision
3
Читать
arXiv Machine Learning
arXiv Machine Learning

Discrete Diffusion Language Models Are Training-Free Multi-Label Classifiers

arXiv:2608.14649v1 Announce Type: new Abstract: We present dLLM-SetScore, a training-free method that uses discrete masked-diffusion language models for multi-label text classification. For each candi...

arXiv Machine Learning
2
Читать
arXiv Computer Vision
arXiv Computer Vision

Concept Guidance: Precise, Training-Free Latent Control for Text-to-Image Generation

arXiv:2608.14172v1 Announce Type: new Abstract: Text-to-image diffusion models have two major drawbacks that severely limit their practical utility: (1) standard models lack an intrinsic mechanism for...

arXiv Computer Vision
4
Читать
arXiv Computer Vision
arXiv Computer Vision

Owner3D: Ownership-Guided Style Writing for Training-Free Localized 3D Stylization

arXiv:2608.14078v1 Announce Type: new Abstract: Localized 3D stylization aims to modify the appearance of a specified object part while preserving the remaining surfaces. In large reconstruction model...

arXiv Computer Vision
2
Читать
arXiv Computer Vision
arXiv Computer Vision

PROVE: Training-Free Prompt Recovery using Verifiable Evidence

arXiv:2608.13671v1 Announce Type: new Abstract: Modern text-to-image models can generate highly realistic images from natural-language prompts, while recent advances in prompt inversion have made it i...

arXiv Computer Vision
3
Читать
arXiv Machine Learning
arXiv Machine Learning

Training-Free Knowledge Transfer Across Model Scales through Activation-Guided Pruning

arXiv:2608.13596v1 Announce Type: new Abstract: Heterogeneous model fusion seeks to combine models that differ in tasks, initializations, architectures, or scales. We study an underexplored cross-scal...

arXiv Machine Learning
4
Читать