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.
By Moritz Gr\"avinghoff, Konstantin Wehmeyer, Aron Asefaw, Boris Knyazev, Damian Borth
arXiv:2410. 10137v5 Announce Type: replace Abstract: We develop Riemannian approaches to variational autoencoders (VAEs) for PDE-type ambient data with regularizing geometric latent dynamics, which we refer to as VAE-DLM, or VAEs with dynamical latent manifolds.
By Andrew Gracyk
arXiv:2501. 09876v3 Announce Type: replace-cross Abstract: Generative modeling aims to generate new data samples that resemble a given dataset.
By Wonjun Lee, Riley C. W. O'Neill, Dongmian Zou, Jeff Calder, Gilad Lerman
arXiv:2505. 04486v4 Announce Type: replace-cross Abstract: Flow matching models have shown great potential in image generation tasks among probabilistic generative models.
By Anirban Samaddar, Yixuan Sun, Viktor Nilsson, Sandeep Madireddy
arXiv:2608. 08135v1 Announce Type: cross Abstract: Cross-modality medical image translation can reduce the burden of multi-modal acquisitions, yet the field remains constrained by two coupled limitations: methods operate on 2D slices or 3D patches rather than whole volumes, and train a separate model for each translation task.
By Daniele Molino, Alessio Zoboli, Camillo Maria Caruso, Valerio Guarrasi, Paolo Soda
arXiv:2607. 03551v1 Announce Type: new Abstract: Weight space learning aims to learn representations of neural network (NN) weights, enabling different downstream tasks.
By Aron Asefaw, Konstantinos Tzevelekakis, Damian Falk, L\'eo Meynent, Damian Borth
arXiv:2512. 01759v3 Announce Type: replace Abstract: We investigate the potential of weights to serve as effective representations, focusing on neural fields.
By Zhuoqian Yang, Mathieu Salzmann, Sabine S\"usstrunk
arXiv:2603.16568v2 Announce Type: replace-cross
Abstract: Manipulating the structure of data in the latent space of an autoencoder is a long-standing goal of representation learning. We propose a sim...
By Laurent Cheret, Vincent L\'etourneau, Isar Nejadgholi, Chris Drummond, Hussein Al Osman, Maia Fraser
MeshCarve introduces a flow‑matching approach that generates artisan meshes entirely within compact latent spaces, separating vertex positions and edge connections to avoid joint encoding issues. It employs a hierarchical sparse transformer backbone with VertexVAE and EdgeVAE, which anchor on discrete vertices to drastically shorten token sequences and use spatial‑aware compression without sacrificing reconstruction quality. The method outperforms state‑of‑the‑art autoregressive and flow‑matching techniques on Objaverse and generalizes to Toys4K, representing one of the first fully compressed latent artisan mesh generators.
By Xiyu Wang, Ruocheng Wu, Yufei Wang, Zhihao Li, Lanqing Guo, Bihan Wen
arXiv:2605. 30705v2 Announce Type: replace-cross Abstract: Geometry-aware generative models and novel view synthesis approaches have shown strong potential in visual fidelity and consistency.
By Sunghyun Kim, Jaehoon Hahm, Jeongwoo Shin, Joonseok Lee
arXiv:2602. 15727v2 Announce Type: replace-cross Abstract: Visual analogy learning enables image editing via demonstration rather than textual description, allowing users to specify complex transformations difficult to articulate in words.
By Hila Manor, Rinon Gal, Haggai Maron, Tomer Michaeli, Gal Chechik
arXiv:2608.29867v1 Announce Type: new
Abstract: Autoencoders are widely used for nonlinear dimensionality reduction and manifold learning. While most common implementations rely on both nonlinear enc...
By Louen Pottier, Louis Lesueur, Anders Thorin