arXiv:2512. 01759v3 Announce Type: replace Abstract: We investigate the potential of weights to serve as effective representations, focusing on neural fields.
arXiv:2607. 03551v1 Announce Type: new Abstract: Weight space learning aims to learn representations of neural network (NN) weights, enabling different downstream tasks.
The paper investigates treating LoRA adapters as a data modality by training a transformer autoencoder to map adapter weights into a learned latent space. It finds that this latent representation preserves the same capabilities as raw weight space—such as sampling, editing, and inversion—while offering greater robustness to displacement, producing distorted yet valid faces instead of texture loss. Additionally, the authors extend the model to accommodate mixed-rank adapters, enabling a single shared latent to host heterogeneous adapter architectures.
The paper investigates treating LoRA adapters as a data modality rather than finished artifacts. By training a transformer autoencoder on LoRA identity adapters, the authors map the adapters into a learned, nonlinear latent space and compare operations such as sampling, editing, and inversion against direct weight-space manipulation. The latent space preserves all capabilities of weight space, offers greater robustness to displacement, and can accommodate mixed-rank adapters within a single shared representation, suggesting a path toward a general, rank‑agnostic adapter modality.
arXiv:2512. 20963v3 Announce Type: replace Abstract: Diffusion models excel at generating high-quality, diverse samples, yet they risk memorizing training data when overfit to the training objective.
arXiv:2610.15571v2 Announce Type: replace Abstract: In this work, we propose a transfer learning algorithm for covariate shift, integrating sample reweighting with representation learning. A key feat...
arXiv:2306.08757v2 Announce Type: replace-cross Abstract: While diffusion models excel at generating high-quality samples, their latent variables typically lack semantic meaning and are not suitable...
arXiv:2501. 09876v3 Announce Type: replace-cross Abstract: Generative modeling aims to generate new data samples that resemble a given dataset.
arXiv:2606.13971v3 Announce Type: replace Abstract: While personalizing Image-to-Video (I2V) diffusion models with specific visual effects is increasingly demanded for high-end generation, current pr...
Recent advances in Diffusion Transformers (DiTs) have enabled remarkable progress in visual synthesis, benefiting from their superior scalability. To facilitate DiTs' capability of capturing meaningful internal representations, recent works such as REPA incorporate external pretrained encoders for representation alignment.
Low-Rank Adaptation (LoRA) has made diffusion model personalization affordable and widely shareable, but it also creates moderation risks when harmful or problematic content is not faithfully disclose...
arXiv:2412.12048v2 Announce Type: replace Abstract: Low-Rank Adaptation (LoRA) has become a standard approach for parameter-efficient fine-tuning of image diffusion models, with over 200K models trai...
Diffu-LoRA is a parameter‑efficient method for personalizing text‑to‑image diffusion models. It inserts trainable low‑rank components into Transformer linear layers and assigns a learnable gate to each component, allowing the model to learn how to distribute adaptation capacity across layers. Using bilevel optimization and progressive pruning, Diffu-LoRA keeps the pretrained backbone frozen while improving subject fidelity and prompt alignment compared to fine‑tuning baselines.