Test-Time Conditioning with Representation-Aligned Visual Features

• URL: http://arxiv.org/abs/2602.03753v1
• Date: Tue, 03 Feb 2026 17:15:03 GMT
• Title: Test-Time Conditioning with Representation-Aligned Visual Features
• Abstract summary: We introduce Representation-Aligned Guidance (REPA-G), a framework that leverages aligned representations with rich semantic properties.<n>We steer the denoising process toward a conditioned representation extracted from a pre-trained feature extractor.<n>Our method provides versatile control at multiple scales, ranging from fine-grained texture matching via single patches to broad semantic guidance.
• Score: 9.262325724962485
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While representation alignment with self-supervised models has been shown to improve diffusion model training, its potential for enhancing inference-time conditioning remains largely unexplored. We introduce Representation-Aligned Guidance (REPA-G), a framework that leverages these aligned representations, with rich semantic properties, to enable test-time conditioning from features in generation. By optimizing a similarity objective (the potential) at inference, we steer the denoising process toward a conditioned representation extracted from a pre-trained feature extractor. Our method provides versatile control at multiple scales, ranging from fine-grained texture matching via single patches to broad semantic guidance using global image feature tokens. We further extend this to multi-concept composition, allowing for the faithful combination of distinct concepts. REPA-G operates entirely at inference time, offering a flexible and precise alternative to often ambiguous text prompts or coarse class labels. We theoretically justify how this guidance enables sampling from the potential-induced tilted distribution. Quantitative results on ImageNet and COCO demonstrate that our approach achieves high-quality, diverse generations. Code is available at https://github.com/valeoai/REPA-G.

Related papers

• Where Does Generative Difficulty Reside? An Empirical Study of Target Representations [62.54876287800644]
  The target representation defines the distribution an image generator must learn, yet it is often treated as an interchangeable interface.<n>We study raw pixels, SD-VAE latents and DINOv2 as well as MAE representation-autoencoder features within a unified masked autoregressive rectified-flow model.<n>Our results show that compression, reconstruction fidelity, token dimensionality, and visible semantic clustering do not individually predict generative behavior.
  arXiv  Detail & Related papers  (2026-08-01T12:32:31Z)
• Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification [80.62512020268626]
  UniAR is a unified autoregressive framework where a single visual tokenizer serves as the key bridge between understanding and generation.<n>UniAR adapts a pretrained vision encoder with multi-level feature fusion and a lookup-free bitwise quantization scheme.<n>A diffusion-based visual decoder operates on discrete visual tokens to decode high-fidelity images.
  arXiv  Detail & Related papers  (2026-06-16T17:59:22Z)
• [CLS] is Not Enough: Multi-Label Recognition via Patch-Level Inference and Adaptive Aggregation [20.637119409165418]
  PIAA formulates prediction as Patch-level Inference followed by Adaptive Aggregation.<n>We introduce an adaptive aggregation module that consolidates patch-level scores into the final multi-label prediction.<n>Experiments show that our method achieves strong improvements with minimal extra computation.
  arXiv  Detail & Related papers  (2026-05-25T13:19:12Z)
• From Sparse to Dense: Multi-View GRPO for Flow Models via Augmented Condition Space [78.36537400975298]
  Group Relative Policy Optimization (GRPO) has emerged as a powerful framework for preference alignment in text-to-image (T2I) flow models.<n>We propose Multi-View GRPO, a novel approach that enhances relationship exploration by augmenting the condition space.<n>MV-GRPO achieves superior alignment performance over state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-03-13T04:35:13Z)
• Training-Free Representation Guidance for Diffusion Models with a Representation Alignment Projector [14.027059904924135]
  We introduce a representation alignment projector that injects representations predicted by a projector into intermediate sampling steps.<n>Experiments on SiTs and REPAs show notable improvements in class-conditional ImageNet synthesis.<n>The proposed method outperforms representative guidance when applied to SiT models.
  arXiv  Detail & Related papers  (2026-01-30T02:29:54Z)
• GloTok: Global Perspective Tokenizer for Image Reconstruction and Generation [51.95701097588426]
  We introduce a Global Perspective Tokenizer (GloTok) to model a more uniform semantic distribution of tokenized features.<n>A residual learning module is proposed to recover the fine-grained details to minimize the reconstruction error caused by quantization.<n>Experiments on the standard ImageNet-1k benchmark clearly show that our proposed method achieves state-of-the-art reconstruction performance and generation quality.
  arXiv  Detail & Related papers  (2025-11-18T06:40:26Z)
• Unsupervised Representation Learning by Balanced Self Attention Matching [2.3020018305241337]
  We present a self-supervised method for embedding image features called BAM. We obtain rich representations and avoid feature collapse by minimizing a loss that matches these distributions to their globally balanced and entropy regularized version. We show competitive performance with leading methods on both semi-supervised and transfer-learning benchmarks.
  arXiv  Detail & Related papers  (2024-08-04T12:52:44Z)
• Readout Guidance: Learning Control from Diffusion Features [96.22155562120231]
  We present Readout Guidance, a method for controlling text-to-image diffusion models with learned signals. Readout Guidance uses readout heads, lightweight networks trained to extract signals from the features of a pre-trained, frozen diffusion model at every timestep. These readouts can encode single-image properties, such as pose, depth, and edges; or higher-order properties that relate multiple images, such as correspondence and appearance similarity.
  arXiv  Detail & Related papers  (2023-12-04T18:59:32Z)
• DiffuseGAE: Controllable and High-fidelity Image Manipulation from Disentangled Representation [14.725538019917625]
  Diffusion probabilistic models (DPMs) have shown remarkable results on various image synthesis tasks. DPMs lack a low-dimensional, interpretable, and well-decoupled latent code. We propose Diff-AE to explore the potential of DPMs for representation learning via autoencoding.
  arXiv  Detail & Related papers  (2023-07-12T04:11:08Z)
• End-to-End Diffusion Latent Optimization Improves Classifier Guidance [81.27364542975235]
  Direct Optimization of Diffusion Latents (DOODL) is a novel guidance method. It enables plug-and-play guidance by optimizing diffusion latents. It outperforms one-step classifier guidance on computational and human evaluation metrics.
  arXiv  Detail & Related papers  (2023-03-23T22:43:52Z)
• Semantic Image Synthesis via Diffusion Models [174.24523061460704]
  Denoising Diffusion Probabilistic Models (DDPMs) have achieved remarkable success in various image generation tasks.<n>Recent work on semantic image synthesis mainly follows the de facto GAN-based approaches.<n>We propose a novel framework based on DDPM for semantic image synthesis.
  arXiv  Detail & Related papers  (2022-06-30T18:31:51Z)
• Diverse Semantic Image Synthesis via Probability Distribution Modeling [103.88931623488088]
  We propose a novel diverse semantic image synthesis framework. Our method can achieve superior diversity and comparable quality compared to state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-03-11T18:59:25Z)
• Unsupervised Discovery of Disentangled Manifolds in GANs [74.24771216154105]
  Interpretable generation process is beneficial to various image editing applications. We propose a framework to discover interpretable directions in the latent space given arbitrary pre-trained generative adversarial networks.
  arXiv  Detail & Related papers  (2020-11-24T02:18:08Z)
• Generalized Adversarially Learned Inference [42.40405470084505]
  We develop methods of inference of latent variables in GANs by adversarially training an image generator along with an encoder to match two joint distributions of image and latent vector pairs. We incorporate multiple layers of feedback on reconstructions, self-supervision, and other forms of supervision based on prior or learned knowledge about the desired solutions.
  arXiv  Detail & Related papers  (2020-06-15T02:18:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.