MC-RFM: Geometry-Aware Few-Shot Adaptation via Mixed-Curvature Riemannian Flow Matching

Hugging Face Daily Papers Papers
few-shot-adaptation geometry-aware riemannian-flow-matching mixed-curvature visual-recognition pretrained-models

Summary

MC-RFM proposes a novel Riemannian flow-matching framework for few-shot adaptation that models feature displacement on a mixed-curvature manifold combining hyperbolic and Euclidean spaces, outperforming existing methods across multiple visual recognition benchmarks.

Parameter-efficient adaptation of pretrained vision models is commonly performed through linear probes, prompts, low-rank updates, or lightweight residual modules. While effective, these methods usually treat adaptation as a discrete Euclidean perturbation of frozen representations, without explicitly modeling the geometry of the task-induced feature displacement. We propose MC-RFM, a mixed-curvature Riemannian flow-matching framework for few-shot adaptation of frozen visual backbones. The key idea is to represent adapted features on a product manifold combining a hyperbolic factor, which captures hierarchy-sensitive semantic structure, and a Euclidean factor, which preserves locally discriminative visual variation. Adaptation is formulated as a task-conditioned continuous transport from frozen features to support-set prototypes, trained with a flow-matching objective and coupled to a hybrid prototype-linear classifier. The method is lightweight, backbone-agnostic, and operates entirely on cached frozen features. Across seven visual recognition benchmarks, five frozen backbones, and 1/4/16-shot regimes, MC-RFM is the best-performing method in a majority of evaluated settings, with the strongest gains on Transformer backbones and fine-grained datasets. Ablations show that the mixed-curvature head, task conditioning, adaptive branch gating, prototype shrinkage, and discriminative supervision each contribute to performance. These results suggest that few-shot adaptation benefits not only from deciding which parameters to update, but also from modeling how representations should move through a geometry matched to the structure of the downstream task.
Original Article
View Cached Full Text

Cached at: 05/15/26, 12:21 AM

Paper page - MC-RFM: Geometry-Aware Few-Shot Adaptation via Mixed-Curvature Riemannian Flow Matching

Source: https://huggingface.co/papers/2605.08557

Abstract

A novel Riemannian flow-matching framework for few-shot adaptation that models feature displacement on a mixed-curvature manifold combining hyperbolic and Euclidean spaces, outperforming existing methods across multiple benchmarks.

Parameter-efficient adaptationof pretrainedvision modelsis commonly performed through linear probes, prompts, low-rank updates, or lightweight residual modules. While effective, these methods usually treat adaptation as a discrete Euclidean perturbation of frozen representations, without explicitly modeling the geometry of the task-inducedfeature displacement. We propose MC-RFM, a mixed-curvatureRiemannian flow-matchingframework forfew-shot adaptationoffrozen visual backbones. The key idea is to represent adapted features on aproduct manifoldcombining ahyperbolic factor, which captures hierarchy-sensitive semantic structure, and aEuclidean factor, which preserves locally discriminative visual variation. Adaptation is formulated as a task-conditioned continuous transport from frozen features to support-set prototypes, trained with aflow-matching objectiveand coupled to ahybrid prototype-linear classifier. The method is lightweight, backbone-agnostic, and operates entirely on cached frozen features. Across seven visual recognition benchmarks, five frozen backbones, and 1/4/16-shot regimes, MC-RFM is the best-performing method in a majority of evaluated settings, with the strongest gains on Transformer backbones and fine-grained datasets. Ablations show that the mixed-curvature head, task conditioning, adaptive branch gating, prototype shrinkage, and discriminative supervision each contribute to performance. These results suggest thatfew-shot adaptationbenefits not only from deciding which parameters to update, but also from modeling how representations should move through a geometry matched to the structure of the downstream task.

View arXiv pageView PDFGitHub4Add to collection

Get this paper in your agent:

hf papers read 2605\.08557

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.08557 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.08557 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.08557 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Geometry-Aware Image Flow Matching

Hugging Face Daily Papers

This paper introduces geometry-aware flow matching for natural images by treating them as points on a hypersphere, proposing SOT-CFM and SFM methods that improve generative modeling by leveraging the spherical structure of image data.

CAT-Flow: Curvature-Adaptive sTeps for Flow Matching

arXiv cs.LG

This paper proposes CAT-OV and CAT-OT, two lightweight, training-free algorithms that adapt step-sizes in Flow Matching sampling based on curvature, improving image quality and reducing generation steps by up to 40%.

Continuous Adversarial MeanFlow Transfer

arXiv cs.LG

This paper proposes MeanFlow-Transfer (MF-T) and Continuous Adversarial MeanFlow (CAMF) to unify the adaptation and acceleration of pretrained diffusion and flow models, enabling high-quality few-step generation on new domains with limited data.