Tootfinder

MOVER: Multimodal Optimal Transport with Volume-based Embedding Regularization
Haochen You, Baojing Liu
https://arxiv.org/abs/2508.12149 https://arxiv.org/…

MOVER: Multimodal Optimal Transport with Volume-based Embedding Regularization
Recent advances in multimodal learning have largely relied on pairwise contrastive objectives to align different modalities, such as text, video, and audio, in a shared embedding space. While effective in bi-modal setups, these approaches struggle to generalize across multiple modalities and often lack semantic structure in high-dimensional spaces. In this paper, we propose MOVER, a novel framework that combines optimal transport-based soft alignment with volume-based geometric regularization t…

Word Meanings in Transformer Language Models
Jumbly Grindrod, Peter Grindrod
https://arxiv.org/abs/2508.12863 https://arxiv.org/pdf/2508.12863

Word Meanings in Transformer Language Models
We investigate how word meanings are represented in the transformer language models. Specifically, we focus on whether transformer models employ something analogous to a lexical store - where each word has an entry that contains semantic information. To do this, we extracted the token embedding space of RoBERTa-base and k-means clustered it into 200 clusters. In our first study, we then manually inspected the resultant clusters to consider whether they are sensitive to semantic information. In …

Weakly Supervised Anomaly Detection in Events with a Higgs Boson and Exotic Physics
Chi Lung Cheng, Sarah Demers, Sascha Diefenbacher, Runze Li, Benjamin Nachman, Dennis Noll
https://arxiv.org/abs/2508.13566

Weakly Supervised Anomaly Detection in Events with a Higgs Boson and Exotic Physics
We present a machine learning-based anomaly detection strategy designed to identify anomalous physics in events containing resonant Standard Model physics and demonstrate this method on the final state of a Higgs boson decaying to two photons. The demonstration targets high-dimensional deviations in the region of phase space containing the Higgs mass peak in a fully signal-agnostic manner. A latent-space embedding, learned from event kinematics, enables the use of a large set of potentially sen…

CASPER: Concept-integrated Sparse Representation for Scientific Retrieval
Lam Thanh Do, Linh Van Nguyen, David Fu, Kevin Chen-Chuan Chang
https://arxiv.org/abs/2508.13394 https:…

CASPER: Concept-integrated Sparse Representation for Scientific Retrieval
The exponential growth of scientific literature has made it increasingly difficult for researchers to keep up with the literature. In an attempt to alleviate this problem, we propose CASPER, a sparse retrieval model for scientific search that utilizes tokens and keyphrases as representation units (i.e. dimensions in the sparse embedding space), enabling it to represent queries and documents with research concepts and match them at both granular and conceptual levels. To overcome the lack of sui…

SAGE: Scale-Aware Gradual Evolution for Continual Knowledge Graph Embedding
Yifei Li, Lingling Zhang, Hang Yan, Tianzhe Zhao, Zihan Ma, Muye Huang, Jun Liu
https://arxiv.org/abs/2508.11347

SAGE: Scale-Aware Gradual Evolution for Continual Knowledge Graph Embedding
Traditional knowledge graph (KG) embedding methods aim to represent entities and relations in a low-dimensional space, primarily focusing on static graphs. However, real-world KGs are dynamically evolving with the constant addition of entities, relations and facts. To address such dynamic nature of KGs, several continual knowledge graph embedding (CKGE) methods have been developed to efficiently update KG embeddings to accommodate new facts while maintaining learned knowledge. As KGs grow at di…

Evaluating Identity Leakage in Speaker De-Identification Systems
Seungmin Seo, Oleg Aulov, Afzal Godil, Kevin Mangold
https://arxiv.org/abs/2508.14012 https://

Evaluating Identity Leakage in Speaker De-Identification Systems
Speaker de-identification aims to conceal a speaker's identity while preserving intelligibility of the underlying speech. We introduce a benchmark that quantifies residual identity leakage with three complementary error rates: equal error rate, cumulative match characteristic hit rate, and embedding-space similarity measured via canonical correlation analysis and Procrustes analysis. Evaluation results reveal that all state-of-the-art speaker de-identification systems leak identity information.…

Targeted Pooled Latent-Space Steganalysis Applied to Generative Steganography, with a Fix
Etienne Levecque (LIST3N), Aur\'elien Noirault (CRIStAL), Tom\'a\v{s} Pevn\'y (CTU), Jan Butora (CRIStAL), Patrick Bas (CRIStAL), R\'emi Cogranne (LIST3N)
https://arxiv.org/abs/2510.12414

Targeted Pooled Latent-Space Steganalysis Applied to Generative Steganography, with a Fix
Steganographic schemes dedicated to generated images modify the seed vector in the latent space to embed a message, whereas most steganalysis methods attempt to detect the embedding in the image space. This paper proposes to perform steganalysis in the latent space by modeling the statistical distribution of the norm of the latent vector. Specifically, we analyze the practical security of a scheme proposed by Hu et. al. for latent diffusion models, which is both robust and practically undetecta…

TopoSizing: An LLM-aided Framework of Topology-based Understanding and Sizing for AMS Circuits
Ziming Wei, Zichen Kong, Yuan Wang, David Z. Pan, Xiyuan Tang
https://arxiv.org/abs/2509.14169

TopoSizing: An LLM-aided Framework of Topology-based Understanding and Sizing for AMS Circuits
Analog and mixed-signal circuit design remains challenging due to the shortage of high-quality data and the difficulty of embedding domain knowledge into automated flows. Traditional black-box optimization achieves sampling efficiency but lacks circuit understanding, which often causes evaluations to be wasted in low-value regions of the design space. In contrast, learning-based methods embed structural knowledge but are case-specific and costly to retrain. Recent attempts with large language m…

Lost in Embeddings: Information Loss in Vision-Language Models
Wenyan Li, Raphael Tang, Chengzu Li, Caiqi Zhang, Ivan Vuli\'c, Anders S{\o}gaard
https://arxiv.org/abs/2509.11986

Lost in Embeddings: Information Loss in Vision-Language Models
Vision--language models (VLMs) often process visual inputs through a pretrained vision encoder, followed by a projection into the language model's embedding space via a connector component. While crucial for modality fusion, the potential information loss induced by this projection step and its direct impact on model capabilities remain understudied. We introduce two complementary approaches to examine and quantify this loss by analyzing the latent representation space. First, we evaluate seman…

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[1/8]:
- Exploring Scholarly Data by Semantic Query on Knowledge Graph Embedding Space
Hung Nghiep Tran, Atsuhiro Takasu

SCDTour: Embedding Axis Ordering and Merging for Interpretable Semantic Change Detection
Taichi Aida, Danushka Bollegala
https://arxiv.org/abs/2509.11818 https://

SCDTour: Embedding Axis Ordering and Merging for Interpretable Semantic Change Detection
In Semantic Change Detection (SCD), it is a common problem to obtain embeddings that are both interpretable and high-performing. However, improving interpretability often leads to a loss in the SCD performance, and vice versa. To address this problem, we propose SCDTour, a method that orders and merges interpretable axes to alleviate the performance degradation of SCD. SCDTour considers both (a) semantic similarity between axes in the embedding space, as well as (b) the degree to which each axi…

SPANER: Shared Prompt Aligner for Multimodal Semantic Representation
Thye Shan Ng, Caren Soyeon Han, Eun-Jung Holden
https://arxiv.org/abs/2508.13387 https://

SPANER: Shared Prompt Aligner for Multimodal Semantic Representation
Recent advances in multimodal Parameter-Efficient Fine-Tuning (PEFT) have significantly improved performance on downstream tasks such as few-shot retrieval. However, most existing approaches focus on task-specific gains while neglecting the structure of the multimodal embedding space. As a result, modality-specific representations often remain isolated, limiting cross-modal generalisation. In this work, we introduce Shared Prompt AligNER (SPANER), a modality-agnostic PEFT framework designed to …

A Lipschitz Refinement of the Multidimensional Bebutov--Kakutani Dynamical Embedding Theorem
Yonatan Gutman, Qiang Huo, Masaki Tsukamoto
https://arxiv.org/abs/2510.08706 https:/…

A Lipschitz Refinement of the Multidimensional Bebutov--Kakutani Dynamical Embedding Theorem
We prove that a continuous action of $\mathbb{R}^n$ on a compact metrizable space equivariantly embeds into the shift action on the space of one-Lipschitz functions from $\mathbb{R}^n$ to $[0,1]$ if and only if the set of fixed points topologically embeds in $[0,1]$. This is a Lipschitz refinement of classical dynamical embedding theorems of Bebutov, Kakutani, Jaworski and Chen.

MA-DPR: Manifold-aware Distance Metrics for Dense Passage Retrieval
Yifan Liu, Qianfeng Wen, Mark Zhao, Jiazhou Liang, Scott Sanner
https://arxiv.org/abs/2509.13562 https://

MA-DPR: Manifold-aware Distance Metrics for Dense Passage Retrieval
Dense Passage Retrieval (DPR) typically relies on Euclidean or cosine distance to measure query-passage relevance in embedding space, which is effective when embeddings lie on a linear manifold. However, our experiments across DPR benchmarks suggest that embeddings often lie on lower-dimensional, non-linear manifolds, especially in out-of-distribution (OOD) settings, where cosine and Euclidean distance fail to capture semantic similarity. To address this limitation, we propose a manifold-aware …

Non-Hermitian Realization of Quantum Dynamics on Embedded Manifolds
Samuel Alperin
https://arxiv.org/abs/2510.11845 https://arxiv.org/pdf/2510.11845…

Non-Hermitian Realization of Quantum Dynamics on Embedded Manifolds
We show that the Floquet Hamiltonian of a quantum particle driven by a general time-periodic imaginary potential is exactly equivalent, at stroboscopic times, to the Hamiltonian of a free particle constrained to a curved Riemannian manifold with fixed embedding. We illustrate the construction for a sinusoidal drive and for the torus of revolution, and outline how the framework can guide experimental design of curved-space quantum dynamics. Our results unify non-Hermitian Floquet physics with sp…

Noise Supervised Contrastive Learning and Feature-Perturbed for Anomalous Sound Detection
Shun Huang, Zhihua Fang, Liang He
https://arxiv.org/abs/2509.13853 https://

Noise Supervised Contrastive Learning and Feature-Perturbed for Anomalous Sound Detection
Unsupervised anomalous sound detection aims to detect unknown anomalous sounds by training a model using only normal audio data. Despite advancements in self-supervised methods, the issue of frequent false alarms when handling samples of the same type from different machines remains unresolved. This paper introduces a novel training technique called one-stage supervised contrastive learning (OS-SCL), which significantly addresses this problem by perturbing features in the embedding space and em…

Optimization of geometric hypergraph embedding
Francesco Zigliotto, Desmond J. Higham
https://arxiv.org/abs/2509.08772 https://arxiv.org/pdf/2509.08772

Optimization of geometric hypergraph embedding
We consider the problem of embedding the nodes of a hypergraph into Euclidean space under the assumption that the interactions arose through closeness to unknown hyperedge centres. In this way, we tackle the inverse problem associated with the generation of geometric random hypergraphs. We propose two new spectral algorithms; both of these exploit the connection between hypergraphs and bipartite graphs. The assumption of an underlying geometric structure allows us to define a concrete measure o…

Image selective encryption analysis using mutual information in CNN based embedding space
Ikram Messadi, Giulia Cervia, Vincent Itier
https://arxiv.org/abs/2508.08832 https://…

Image selective encryption analysis using mutual information in CNN based embedding space
As digital data transmission continues to scale, concerns about privacy grow increasingly urgent - yet privacy remains a socially constructed and ambiguously defined concept, lacking a universally accepted quantitative measure. This work examines information leakage in image data, a domain where information-theoretic guarantees are still underexplored. At the intersection of deep learning, information theory, and cryptography, we investigate the use of mutual information (MI) estimators - in pa…

Compressibility Barriers to Neighborhood-Preserving Data Visualizations
Szymon Snoeck, Noah Bergam, Nakul Verma
https://arxiv.org/abs/2508.07119 https://ar…

Compressibility Barriers to Neighborhood-Preserving Data Visualizations
To what extent is it possible to visualize high-dimensional datasets in a two- or three-dimensional space? We reframe this question in terms of embedding $n$-vertex graphs (representing the neighborhood structure of the input points) into metric spaces of low doubling dimension $d$, in such a way that maintains the separation between neighbors and non-neighbors. This seemingly lax embedding requirement is surprisingly difficult to satisfy. Our investigation shows that an overwhelming fraction o…

Quantum-inspired probability metrics define a complete, universal space for statistical learning
Logan S. McCarty
https://arxiv.org/abs/2508.21086 https://…

Quantum-inspired probability metrics define a complete, universal space for statistical learning
Comparing probability distributions is a core challenge across the natural, social, and computational sciences. Existing methods, such as Maximum Mean Discrepancy (MMD), struggle in high-dimensional and non-compact domains. Here we introduce quantum probability metrics (QPMs), derived by embedding probability measures in the space of quantum states: positive, unit-trace operators on a Hilbert space. This construction extends kernel-based methods and overcomes the incompleteness of MMD on non-co…

Singularity of Cannon-Thurston maps
Vaibhav Gadre, Joseph Maher, Catherine Pfaff, Caglar Uyanik
https://arxiv.org/abs/2510.04350 https://arxiv.org/pdf/2510…

Singularity of Cannon-Thurston maps
In a closed fibered hyperbolic 3-manifold M, the inclusion of a fiber S, with S and M lifted to the universal covers gives an exponentially distorted embedding of the hyperbolic plane into hyperbolic 3-space. Nevertheless, Cannon and Thurston showed that there is a map from the circle at infinity of the hyperbolic plane to the 2-sphere at infinity of hyperbolic 3-space. The Cannon-Thurston map is surjective, finite-to-one, and gives a space-filling curve. Here we prove that many natural measure…

Local Density-Based Anomaly Score Normalization for Domain Generalization
Kevin Wilkinghoff, Haici Yang, Janek Ebbers, Fran\c{c}ois G. Germain, Gordon Wichern, Jonathan Le Roux
https://arxiv.org/abs/2509.10951

Local Density-Based Anomaly Score Normalization for Domain Generalization
State-of-the-art anomalous sound detection (ASD) systems in domain-shifted conditions rely on projecting audio signals into an embedding space and using distance-based outlier detection to compute anomaly scores. One of the major difficulties to overcome is the so-called domain mismatch between the anomaly score distributions of a source domain and a target domain that differ acoustically and in terms of the amount of training data provided. A decision threshold that is optimal for one domain m…

Directed Evolution of Proteins via Bayesian Optimization in Embedding Space
Matou\v{s} Sold\'at, Ji\v{r}\'i Kl\'ema
https://arxiv.org/abs/2509.04998 https://

Directed Evolution of Proteins via Bayesian Optimization in Embedding Space
Directed evolution is an iterative laboratory process of designing proteins with improved function by iteratively synthesizing new protein variants and evaluating their desired property with expensive and time-consuming biochemical screening. Machine learning methods can help select informative or promising variants for screening to increase their quality and reduce the amount of necessary screening. In this paper, we present a novel method for machine-learning-assisted directed evolution of pr…

Neural Receptive Fields, Stimulus Space Embedding and Effective Geometry of Scale-Free Networks
Vasilii Tiselko, Alexander Gorsky, Yuri Dabaghian
https://arxiv.org/abs/2509.25453

Neural Receptive Fields, Stimulus Space Embedding and Effective Geometry of Scale-Free Networks
Understanding how neuronal dynamics couple with stimuli space and how receptive fields emerge and organize within brain networks remains a fundamental challenge in neuroscience. Several models attempted to explain these phenomena, often by adjusting the network to empirical manifestations, but struggled to achieve biological plausibility. Here, we propose a physiologically grounded model in which receptive fields and population-level attractor dynamics emerge naturally from the effective hyperb…

Linear codes arising from the point-hyperplane geometry -- Part II: the twisted embedding
Ilaria Cardinali, Luca Giuzzi
https://arxiv.org/abs/2507.16694 ht…

Linear codes arising from the point-hyperplane geometry -- Part II: the twisted embedding
Let $\barΓ$ be the point-hyperplane geometry of a projective space $\mathrm{PG(V)},$ where $V$ is a $(n+1)$-dimensional vector space over a finite field $\mathbb{F}_q$ of order $q.$ Suppose that $σ$ is an automorphism of $\mathbb{F}_q$ and consider the projective embedding $\varepsilon_σ$ of $\barΓ$ into the projective space $\mathrm{PG}(V\otimes V^*)$ mapping the point $([x],[ξ])\in \barΓ$ to the projective point represented by the pure tensor $x^σ\otimes ξ$, with $ξ(x)=0.$ In [I. Car…

Benchmarking Filtered Approximate Nearest Neighbor Search Algorithms on Transformer-based Embedding Vectors
Patrick Iff, Paul Bruegger, Marcin Chrapek, Maciej Besta, Torsten Hoefler
https://arxiv.org/abs/2507.21989

Benchmarking Filtered Approximate Nearest Neighbor Search Algorithms on Transformer-based Embedding Vectors
Advances in embedding models for text, image, audio, and video drive progress across multiple domains, including retrieval-augmented generation, recommendation systems, vehicle/person reidentification, and face recognition. Many applications in these domains require an efficient method to retrieve items that are close to a given query in the embedding space while satisfying a filter condition based on the item's attributes, a problem known as Filtered Approximate Nearest Neighbor Search (FANNS)…

Cauchy problems for time-space fractional coupled chemotaxis-fluid equations in Besov-Morrey spaces
Yong Zhen Yang, Yong Zhou, Xiao Lin Liu
https://arxiv.org/abs/2508.01369 http…

Cauchy problems for time-space fractional coupled chemotaxis-fluid equations in Besov-Morrey spaces
In this paper, we consider the Cauchy problems for the time-space fractional coupled chemotaxis-fluid equations, which is a generalized form of the coupled chemotaxis-fluid equations studied in \cite{M.H. Yang}. In contrast to \cite{M.H. Yang}, the solution operator of the system does not satisfy the semigroup effect, which makes the approach of \cite{M.H. Yang} inapplicable. Based on the theory of harmonic analysis, using techniques such as real interpolation, embedding in Besov-Morrey spaces,…

Exploring the Stratified Space Structure of an RL Game with the Volume Growth Transform
Justin Curry, Brennan Lagasse, Ngoc B. Lam, Gregory Cox, David Rosenbluth, Alberto Speranzon
https://arxiv.org/abs/2507.22010

Exploring the Stratified Space Structure of an RL Game with the Volume Growth Transform
In this work, we explore the structure of the embedding space of a transformer model trained for playing a particular reinforcement learning (RL) game. Specifically, we investigate how a transformer-based Proximal Policy Optimization (PPO) model embeds visual inputs in a simple environment where an agent must collect "coins" while avoiding dynamic obstacles consisting of "spotlights." By adapting Robinson et al.'s study of the volume growth transform for LLMs to the RL setting, we find that the…

FSP-DETR: Few-Shot Prototypical Parasitic Ova Detection
Shubham Trehan, Udhav Ramachandran, Akash Rao, Ruth Scimeca, Sathyanarayanan N. Aakur
https://arxiv.org/abs/2510.09583 ht…

FSP-DETR: Few-Shot Prototypical Parasitic Ova Detection
Object detection in biomedical settings is fundamentally constrained by the scarcity of labeled data and the frequent emergence of novel or rare categories. We present FSP-DETR, a unified detection framework that enables robust few-shot detection, open-set recognition, and generalization to unseen biomedical tasks within a single model. Built upon a class-agnostic DETR backbone, our approach constructs class prototypes from original support images and learns an embedding space using augmented v…

Cross-device Zero-shot Label Transfer via Alignment of Time Series Foundation Model Embeddings
Neal G. Ravindra, Arijit Sehanobish
https://arxiv.org/abs/2509.06966 https://

Cross-device Zero-shot Label Transfer via Alignment of Time Series Foundation Model Embeddings
High-quality, medically validated labels exist for clinical actigraphy data but not for ubiquitous consumer wearables like the Apple Watch. Manually labeling wearables data is expensive and doesn't scale. This paper offers a novel framework that transfers valuable labels from a source domain (e.g., actigraphy) to a target domain (e.g., Apple Watch) without requiring paired data. Instead of working with raw time-series signals, we project both domains into a shared latent embedding space using t…

Secant loci on moduli of Prym varieties
Gavril Farkas, Margherita Lelli-Chiesa
https://arxiv.org/abs/2509.26118 https://arxiv.org/pdf/2509.26118

Secant loci on moduli of Prym varieties
We present a Prym analogue of Lazarsfeld's result that curves on general polarized K3 surfaces verify the Brill-Noether Theorem, or equivalently, that their canonical embedding has no unexpected secants. We show that the Prym-canonical embedding of a curve on a general Nikulin surface (both of standard and non-standard types) has no unexpected secants. We then explain how these geometric facts suffice to determine the class of the universal difference divisor on the moduli space of stable Prym …

Symmetric Spaces with Rectangular Unit Lattices, Revisited
Jost-Hinrich Eschenburg, Ernst Heintze, Peter Quast
https://arxiv.org/abs/2509.16166 https://arx…

Symmetric Spaces with Rectangular Unit Lattices, Revisited
We give a new proof of a theorem of Loos stating that a Riemannian symmetric space X with rectangular unit lattice is a symmetric R-space. For this we construct explicitly an isometric extrinsically symmetric embedding of X in a Euclidean space which reveals X as a standardly embedded symmetric R-space. We further determine the root systems, Euclidean root data, fundamental groups and eigenvalues of the Laplacian of symmetric spaces with rectangular unit lattice in a direct way.

Topolow: Force-Directed Euclidean Embedding of Dissimilarity Data with Robustness Against Non-Metricity and Sparsity
Omid Arhami, Pejman Rohani
https://arxiv.org/abs/2508.01733 …

Topolow: Force-Directed Euclidean Embedding of Dissimilarity Data with Robustness Against Non-Metricity and Sparsity
The problem of embedding a set of objects into a low-dimensional Euclidean space based on a matrix of pairwise dissimilarities is fundamental in data analysis, machine learning, and statistics. However, the assumptions of many standard analytical methods are violated when the input dissimilarities fail to satisfy metric or Euclidean axioms. We present the mathematical and statistical foundations of Topolow, a physics-inspired, gradient-free optimization framework for such embedding problems. To…

Embedding separable Banach spaces into $(\ell^\infty \setminus c)\cup\{0\}
Geivison Ribeiro
https://arxiv.org/abs/2508.18656 https://arxiv.org/pdf/2508.186…

Embedding separable Banach spaces into $(\ell^\infty \setminus c)\cup\{0\}
We prove that every separable Banach space admits an isometric embedding into $\linf$ intersecting $\cseq$ only at the origin. Moreover, we prove that any finite-or countable-dimensional, or more generally separable, subspace of $(\ell_\infty \setminus \cseq)\cup \{0\}$ can be extended to a subspace containing an isometric copy of an arbitrary separable Banach space, while still avoiding $\cseq$.

Scaling Language-Centric Omnimodal Representation Learning
Chenghao Xiao, Hou Pong Chan, Hao Zhang, Weiwen Xu, Mahani Aljunied, Yu Rong
https://arxiv.org/abs/2510.11693 https://…

Scaling Language-Centric Omnimodal Representation Learning
Recent multimodal embedding approaches leveraging multimodal large language models (MLLMs) fine-tuned with contrastive learning (CL) have shown promising results, yet the underlying reasons behind their superiority remain underexplored. This work argues that a crucial advantage of MLLM-based approaches stems from implicit cross-modal alignment achieved during generative pretraining, where the language decoder learns to exploit multimodal signals within a shared representation space for generati…

On Sets of Periodic Orbit Lengths in Finitely Presented Dynamical Systems
Huub de Jong
https://arxiv.org/abs/2510.10848 https://arxiv.org/pdf/2510.10848

On Sets of Periodic Orbit Lengths in Finitely Presented Dynamical Systems
We classify the sets of natural numbers $n$ for which certain dynamical systems $(X,f)$ on a compact metric space $X$ have a periodic point of (least) period $n$. Interest in this question dates back to Sharkovskii's theorem for continuous maps on intervals of the real line, but it also ties to checkable conditions for Krieger's embedding theorem for symbolic dynamical systems. Given a system for which the logarithmic derivative of the Artin-Mazur zeta function is rational, we use the Skolem-Ma…

Convex real projective structures on Coxeter orbifold $D^2(;n_1,n_2,n_3,n_4)\times \mathbb{R}$
Jaesung Bae
https://arxiv.org/abs/2509.05579 https://arxiv.o…

Convex real projective structures on Coxeter orbifold $D^2(;n_1,n_2,n_3,n_4)\times \mathbb{R}$
The deformation space of real projective structures parametrizes the space of the convex real projective structures on an orbifold. The Coxeter orbifold can be obtained $D^2(;n_1,n_2,n_3,n_4)\times\mathbb{R}$ by embedding the Coxeter quadrilateral from $\mathbb{RP}^2$ to $\mathbb{RP}^3$, and extending the side edges to planes and perturbing these to obtain a convex polytope. This noncompact orbifold can be induced from two combinatorial polytopes. Using this fact, we determine the deformation s…

Koopman Operator Based Time-Delay Embeddings and State History Augmented LQR for Periodic Hybrid Systems: Bouncing Pendulum and Bipedal Walking
Chun-Ming Yang, Pranav A. Bhounsule
https://arxiv.org/abs/2507.14455

Koopman Operator Based Time-Delay Embeddings and State History Augmented LQR for Periodic Hybrid Systems: Bouncing Pendulum and Bipedal Walking
Time-delay embedding is a technique that uses snapshots of state history over time to build a linear state space model of a nonlinear smooth system. We demonstrate that periodic non-smooth or hybrid system can also be modeled as a linear state space system using this approach as long as its behavior is consistent in modes and timings. We extended time-delay embeddings to generate a linear model of two periodic hybrid systems: the bouncing pendulum and the simplest walker with control inputs. Th…

Crosslisted article(s) found for q-bio.BM. https://arxiv.org/list/q-bio.BM/new
[1/1]:
- Directed Evolution of Proteins via Bayesian Optimization in Embedding Space
Matou\v{s} Sold\'at, Ji\v{r}\'i Kl\'ema

Mixture of Encoders is a vector-native alternative that models both structured and unstructured data in a unified embedding space. At Berlin Buzzwords 2025, Filip Makraduli and Ben Gutkovich introduced the method, demonstrated how it powers natural language search and real-time recommendations, and shared open-source tools and benchmarks for replacing complex hybrid stacks.
Watch the full session: https://youtu.be/v-ak-Ohjqas?si=Lyvqcr0fb-0sa4ZW
Berlin Buzzwords returns on 7-9 June 2026: https://tickets.plainschwarz.com/bbuzz26/c/8Hvk0ZvJA/

DeePAQ: A Perceptual Audio Quality Metric Based On Foundational Models and Weakly Supervised Learning
Guanxin Jiang, Andreas Brendel, Pablo M. Delgado, J\"urgen Herre
https://arxiv.org/abs/2510.12326

DeePAQ: A Perceptual Audio Quality Metric Based On Foundational Models and Weakly Supervised Learning
This paper presents the Deep learning-based Perceptual Audio Quality metric (DeePAQ) for evaluating general audio quality. Our approach leverages metric learning together with the music foundation model MERT, guided by surrogate labels, to construct an embedding space that captures distortion intensity in general audio. To the best of our knowledge, DeePAQ is the first in the general audio quality domain to leverage weakly supervised labels and metric learning for fine-tuning a music foundation…

Med-GRIM: Enhanced Zero-Shot Medical VQA using prompt-embedded Multimodal Graph RAG
Rakesh Raj Madavan, Akshat Kaimal, Hashim Faisal, Chandrakala S
https://arxiv.org/abs/2508.06496

Med-GRIM: Enhanced Zero-Shot Medical VQA using prompt-embedded Multimodal Graph RAG
An ensemble of trained multimodal encoders and vision-language models (VLMs) has become a standard approach for visual question answering (VQA) tasks. However, such models often fail to produce responses with the detailed precision necessary for complex, domain-specific applications such as medical VQA. Our representation model, BIND: BLIVA Integrated with Dense Encoding, extends prior multimodal work by refining the joint embedding space through dense, query-token-based encodings inspired by c…

EmbedGrad: Gradient-Based Prompt Optimization in Embedding Space for Large Language Models
Xiaoming Hou, Jiquan Zhang, Zibin Lin, DaCheng Tao, Shengli Zhang
https://arxiv.org/abs/2508.03533

EmbedGrad: Gradient-Based Prompt Optimization in Embedding Space for Large Language Models
Effectively adapting powerful pretrained foundation models to diverse tasks remains a key challenge in AI deployment. Current approaches primarily follow two paradigms:discrete optimization of text prompts through prompt engineering, or continuous adaptation via additional trainable parameters. Both exhibit limitations-discrete methods lack refinement precision while parameter-based techniques increase complexity and reduce interpretability. To address these constraints, we propose EmbedGrad, a…

Robust State-space Reconstruction of Brain Dynamics via Bootstrap Monte Carlo SSA
Sir-Lord Wiafe, Carter Hinsley, Vince D. Calhoun
https://arxiv.org/abs/2510.00011 https://

Robust State-space Reconstruction of Brain Dynamics via Bootstrap Monte Carlo SSA
Reconstructing latent state-space geometry from time series provides a powerful route to studying nonlinear dynamics across complex systems. Delay-coordinate embedding provides the theoretical basis but assumes long, noise-free recordings, which many domains violate. In neuroimaging, for example, fMRI is short and noisy; low sampling and strong red noise obscure oscillations and destabilize embeddings. We propose bootstrap Monte Carlo SSA with a red-noise null and bootstrap stability to retain …

Probability Signature: Bridging Data Semantics and Embedding Structure in Language Models
Junjie Yao, Zhi-Qin John Xu
https://arxiv.org/abs/2509.20124 https://

Probability Signature: Bridging Data Semantics and Embedding Structure in Language Models
The embedding space of language models is widely believed to capture the semantic relationships; for instance, embeddings of digits often exhibit an ordered structure that corresponds to their natural sequence. However, the mechanisms driving the formation of such structures remain poorly understood. In this work, we interpret the embedding structures via the data distribution. We propose a set of probability signatures that reflect the semantic relationships among tokens. Through experiments o…

Geodesic complexity of the octahedron, and an algorithm for cut loci on convex polyhedra
Florian Frick, Pranav Rajbhandari
https://arxiv.org/abs/2508.19362 https://

Geodesic complexity of the octahedron, and an algorithm for cut loci on convex polyhedra
The geodesic complexity of a length space $X$ quantifies the required number of case distinctions to continuously choose a shortest path connecting any given start and end point. We prove a local lower bound for the geodesic complexity of $X$ obtained by embedding simplices into $X\times X$. We additionally create and prove correctness of an algorithm to find cut loci on surfaces of convex polyhedra, as the structure of a space's cut loci is related to its geodesic complexity. We use these tech…

Embedding calculus and Vassiliev spectral sequence
Syunji Moriya
https://arxiv.org/abs/2509.23766 https://arxiv.org/pdf/2509.23766

Embedding calculus and Vassiliev spectral sequence
Vassiliev spectral sequence and Sinha spectral sequence are both related to cohomology of the space of long knots $\mathbb{R}\to \mathbb{R}^3$. Although they have different origins, the Vassiliev $E_1$-page and the Sinha $E_2$-page are isomorphic (up to a degree shift). In this paper, we prove that they have isomorphic $E_\infty$-pages if the coefficient ring is a field. Together with degeneracy of the Sinha sequence, this implies that the Vassiliev sequence degenerates at $E_1$-page over $\mat…

The Intrinsic Riemannian Proximal Gradient Method for Convex Optimization
Ronny Bergmann, Hajg Jasa, Paula John, Max Pfeffer
https://arxiv.org/abs/2507.16055

The Intrinsic Riemannian Proximal Gradient Method for Convex Optimization
We consider a class of (possibly strongly) geodesically convex optimization problems on Hadamard manifolds, where the objective function splits into the sum of a smooth and a possibly nonsmooth function. We introduce an intrinsic convex Riemannian proximal gradient (CRPG) method that employs the manifold proximal map for the nonsmooth step, without operating in the embedding or tangent space. A sublinear convergence rate for convex problems and a linear convergence rate for strongly convex prob…

Interpreting CFD Surrogates through Sparse Autoencoders
Yeping Hu, Shusen Liu
https://arxiv.org/abs/2507.16069 https://arxiv.org/pdf/2507.16069

Interpreting CFD Surrogates through Sparse Autoencoders
Learning-based surrogate models have become a practical alternative to high-fidelity CFD solvers, but their latent representations remain opaque and hinder adoption in safety-critical or regulation-bound settings. This work introduces a posthoc interpretability framework for graph-based surrogate models used in computational fluid dynamics (CFD) by leveraging sparse autoencoders (SAEs). By obtaining an overcomplete basis in the node embedding space of a pretrained surrogate, the method extracts…

Projective models for Hilbert squares of $K3$ surfaces
\'Angel David R\'ios Ortiz, Andr\'es Rojas, Jieao Song
https://arxiv.org/abs/2510.02065 https://

Projective models for Hilbert squares of $K3$ surfaces
For a very general polarized $K3$ surface $S\subset \mathbb{P}^g$ of genus $g\ge 5$, we study the linear system on the Hilbert square $S^{[2]}$ parametrizing quadrics in $\mathbb{P}^g$ that contain $S$. We prove its very ampleness for $g\geq 7$. In the cases of genus 7 or 8, we describe in detail the projective geometry of the corresponding embedding by making use of the Mukai model for $S$. In both cases, it can be realized as a degeneracy locus on an ambient homogeneous space, in a strikingly…

Neural Proxies for Sound Synthesizers: Learning Perceptually Informed Preset Representations
Paolo Combes, Stefan Weinzierl, Klaus Obermayer
https://arxiv.org/abs/2509.07635 htt…

Neural Proxies for Sound Synthesizers: Learning Perceptually Informed Preset Representations
Deep learning appears as an appealing solution for Automatic Synthesizer Programming (ASP), which aims to assist musicians and sound designers in programming sound synthesizers. However, integrating software synthesizers into training pipelines is challenging due to their potential non-differentiability. This work tackles this challenge by introducing a method to approximate arbitrary synthesizers. Specifically, we train a neural network to map synthesizer presets onto an audio embedding space …

Transform Before You Query: A Privacy-Preserving Approach for Vector Retrieval with Embedding Space Alignment
Ruiqi He, Zekun Fei, Jiaqi Li, Xinyuan Zhu, Biao Yi, Siyi Lv, Weijie Liu, Zheli Liu
https://arxiv.org/abs/2507.18518

Transform Before You Query: A Privacy-Preserving Approach for Vector Retrieval with Embedding Space Alignment
Vector Database (VDB) can efficiently index and search high-dimensional vector embeddings from unstructured data, crucially enabling fast semantic similarity search essential for modern AI applications like generative AI and recommendation systems. Since current VDB service providers predominantly use proprietary black-box models, users are forced to expose raw query text to them via API in exchange for the vector retrieval services. Consequently, if query text involves confidential records fro…

Language Models Do Not Embed Numbers Continuously
Alex O. Davies, Roussel Nzoyem, Nirav Ajmeri, Telmo M. Silva Filho
https://arxiv.org/abs/2510.08009 https://

Language Models Do Not Embed Numbers Continuously
Recent research has extensively studied how large language models manipulate integers in specific arithmetic tasks, and on a more fundamental level, how they represent numeric values. These previous works have found that language model embeddings can be used to reconstruct the original values, however, they do not evaluate whether language models actually model continuous values as continuous. Using expected properties of the embedding space, including linear reconstruction and principal compon…

Improving Cram\'er-Rao Bound And Its Variants: An Extrinsic Geometry Perspective
Sunder Ram Krishnan
https://arxiv.org/abs/2509.17886 https://arxiv.org…

Improving Cramér-Rao Bound And Its Variants: An Extrinsic Geometry Perspective
This work presents a geometric refinement of the classical Cramér-Rao bound (CRB) by incorporating curvature-aware corrections based on the second fundamental form associated with the statistical model manifold. That is, our formulation shows that relying on the extrinsic geometry of the square root embedding of the manifold in the ambient Hilbert space comprising square integrable functions with respect to a fixed base measure offers a rigorous (and intuitive) way to improve upon the CRB and …

Interpolating Speaker Identities in Embedding Space for Data Expansion
Tianchi Liu, Ruijie Tao, Qiongqiong Wang, Yidi Jiang, Hardik B. Sailor, Ke Zhang, Jingru Lin, Haizhou Li
https://arxiv.org/abs/2508.19210

Interpolating Speaker Identities in Embedding Space for Data Expansion
The success of deep learning-based speaker verification systems is largely attributed to access to large-scale and diverse speaker identity data. However, collecting data from more identities is expensive, challenging, and often limited by privacy concerns. To address this limitation, we propose INSIDE (Interpolating Speaker Identities in Embedding Space), a novel data expansion method that synthesizes new speaker identities by interpolating between existing speaker embeddings. Specifically, we…

A remark on equivariant Riemannian isometric embeddings preserving symmetries
Dmitri Burago, Hongda Qiu
https://arxiv.org/abs/2507.23164 https://arxiv.org/…

A remark on equivariant Riemannian isometric embeddings preserving symmetries
This remark pertains to isometric embeddings endowed with certain geometric properties. We study two embeddings problems for the universal cover $M$ of an $n$-dimensional Riemannian torus $(\TT^n,g)$. The first concerns the existence of an isometric embedding of $M$ into a bounded subset of some Euclidean space $\RR^{D_1}$, and the second one seeks an isometric embdding of $M$ that is equivariant with respect to the deck transformation group of covering map. By using a known trick in a novel wa…

Tac2Motion: Contact-Aware Reinforcement Learning with Tactile Feedback for Robotic Hand Manipulation
Yitaek Kim, Casper Hewson Rask, Christoffer Sloth
https://arxiv.org/abs/2509.17812

Tac2Motion: Contact-Aware Reinforcement Learning with Tactile Feedback for Robotic Hand Manipulation
This paper proposes Tac2Motion, a contact-aware reinforcement learning framework to facilitate the learning of contact-rich in-hand manipulation tasks, such as removing a lid. To this end, we propose tactile sensing-based reward shaping and incorporate the sensing into the observation space through embedding. The designed rewards encourage an agent to ensure firm grasping and smooth finger gaiting at the same time, leading to higher data efficiency and robust performance compared to the baselin…

Kernel K-means clustering of distributional data
Amparo Ba\'illo, Jose R. Berrendero, Mart\'in S\'anchez-Signorini
https://arxiv.org/abs/2509.18037 https://

Kernel K-means clustering of distributional data
We consider the problem of clustering a sample of probability distributions from a random distribution on $\mathbb R^p$. Our proposed partitioning method makes use of a symmetric, positive-definite kernel $k$ and its associated reproducing kernel Hilbert space (RKHS) $\mathcal H$. By mapping each distribution to its corresponding kernel mean embedding in $\mathcal H$, we obtain a sample in this RKHS where we carry out the $K$-means clustering procedure, which provides an unsupervised classifica…

An Ambit Field Framework for the Full Panel of Day-ahead Electricity Prices
Thomas K. Kloster
https://arxiv.org/abs/2509.17236 https://arxiv.org/pdf/2509.1…

An Ambit Field Framework for the Full Panel of Day-ahead Electricity Prices
This paper considers the often overlooked fact that electricity spot prices in individual European generation zones evolve as a high dimensional panel structure. A general continuous time framework is developed by formulating the panel as an ambit field indexed by a cylinder surface, where the cross sectional dimension is represented by a circle. This requires a treatment of ambit fields on manifolds, but the departure from Euclidean space allows for embedding intrinsic dependence structures in…

Besov spaces associated with the Harmonic oscillator
Reika Fukuizumi, Tsukasa Iwabuchi
https://arxiv.org/abs/2508.20436 https://arxiv.org/pdf/2508.20436

Besov spaces associated with the Harmonic oscillator
The Besov space associated with the harmonic oscillator is introduced and thoroughly explored in this paper. It provides a comprehensive summary of the fundamental concepts of the Besov spaces, their embedding properties, bilinear estimates, and related topics.

Parameter-Efficient Single Collaborative Branch for Recommendation
Marta Moscati, Shah Nawaz, Markus Schedl
https://arxiv.org/abs/2508.03518 https://arxiv.…

Parameter-Efficient Single Collaborative Branch for Recommendation
Recommender Systems (RS) often rely on representations of users and items in a joint embedding space and on a similarity metric to compute relevance scores. In modern RS, the modules to obtain user and item representations consist of two distinct and separate neural networks (NN). In multimodal representation learning, weight sharing has been proven effective in reducing the distance between multiple modalities of a same item. Inspired by these approaches, we propose a novel RS that leverages w…

Lightweight and Generalizable Acoustic Scene Representations via Contrastive Fine-Tuning and Distillation
Kuang Yuan, Yang Gao, Xilin Li, Xinhao Mei, Syavosh Zadissa, Tarun Pruthi, Saeed Bagheri Sereshki
https://arxiv.org/abs/2510.03728

Lightweight and Generalizable Acoustic Scene Representations via Contrastive Fine-Tuning and Distillation
Acoustic scene classification (ASC) models on edge devices typically operate under fixed class assumptions, lacking the transferability needed for real-world applications that require adaptation to new or refined acoustic categories. We propose ContrastASC, which learns generalizable acoustic scene representations by structuring the embedding space to preserve semantic relationships between scenes, enabling adaptation to unseen categories without retraining. Our approach combines supervised con…

MLSD: A Novel Few-Shot Learning Approach to Enhance Cross-Target and Cross-Domain Stance Detection
Parush Gera, Tempestt Neal
https://arxiv.org/abs/2509.03725 https://

MLSD: A Novel Few-Shot Learning Approach to Enhance Cross-Target and Cross-Domain Stance Detection
We present the novel approach for stance detection across domains and targets, Metric Learning-Based Few-Shot Learning for Cross-Target and Cross-Domain Stance Detection (MLSD). MLSD utilizes metric learning with triplet loss to capture semantic similarities and differences between stance targets, enhancing domain adaptation. By constructing a discriminative embedding space, MLSD allows a cross-target or cross-domain stance detection model to acquire useful examples from new target domains. We …

Improving Generalizability and Undetectability for Targeted Adversarial Attacks on Multimodal Pre-trained Models
Zhifang Zhang, Jiahan Zhang, Shengjie Zhou, Qi Wei, Shuo He, Feng Liu, Lei Feng
https://arxiv.org/abs/2509.19994

Improving Generalizability and Undetectability for Targeted Adversarial Attacks on Multimodal Pre-trained Models
Multimodal pre-trained models (e.g., ImageBind), which align distinct data modalities into a shared embedding space, have shown remarkable success across downstream tasks. However, their increasing adoption raises serious security concerns, especially regarding targeted adversarial attacks. In this paper, we show that existing targeted adversarial attacks on multimodal pre-trained models still have limitations in two aspects: generalizability and undetectability. Specifically, the crafted targe…

LAVA: Explainability for Unsupervised Latent Embeddings
Ivan Stresec, Joana P. Gon\c{c}alves
https://arxiv.org/abs/2509.21149 https://arxiv.org/pdf/2509.21…

LAVA: Explainability for Unsupervised Latent Embeddings
Unsupervised black-box models can be drivers of scientific discovery, but remain difficult to interpret. Crucially, discovery hinges on understanding the model output, which is often a multi-dimensional latent embedding rather than a well-defined target. While explainability for supervised learning usually seeks to uncover how input features are used to predict a target, its unsupervised counterpart should relate input features to the structure of the learned latent space. Adaptations of superv…

Vertex-minimal hyperbolic origami 2-torus
Zhengyu Zou
https://arxiv.org/abs/2509.18668 https://arxiv.org/pdf/2509.18668

Vertex-minimal hyperbolic origami 2-torus
In this paper, we explore geometrizations of triangulated hyperbolic 2-tori $Σ_2$ in the hyperbolic 3-space $\mathbb{H}^3$. We show that for all $n \geq 10$, there exists a triangulation of $Σ_2$ with $n$ vertices and an embedding $S: Σ_2 \hookrightarrow \mathbb{H}^3$ that restricts to isometries on the triangles. We call these embeddings hyperbolic origami 2-tori. Since the lower bound $n \geq 10$ is combinatorially sharp, this paper settles the question of minimum number of vertices requir…

Spectral representations of interpolation spaces of reproducing kernel Hilbert spaces
Michael Bitzer, Ingo Steinwart
https://arxiv.org/abs/2508.16492 https://

Spectral representations of interpolation spaces of reproducing kernel Hilbert spaces
In statistical learning theory, interpolation spaces of the form $[\mathrm{L}^2,H]_{θ,r}$, where $H$ is a reproducing kernel Hilbert space, are in widespread use. So far, however, they are only well understood for fine index $r=2$. We generalise existing results from $r=2$ to all possible values of $r$. In particular, we present a spectral decomposition of such spaces, analyse their embedding properties, and describe connections to the theory of Banach spaces of functions. We additionally pres…

SIDA: Synthetic Image Driven Zero-shot Domain Adaptation
Ye-Chan Kim, SeungJu Cha, Si-Woo Kim, Taewhan Kim, Dong-Jin Kim
https://arxiv.org/abs/2507.18632 https://

SIDA: Synthetic Image Driven Zero-shot Domain Adaptation
Zero-shot domain adaptation is a method for adapting a model to a target domain without utilizing target domain image data. To enable adaptation without target images, existing studies utilize CLIP's embedding space and text description to simulate target-like style features. Despite the previous achievements in zero-shot domain adaptation, we observe that these text-driven methods struggle to capture complex real-world variations and significantly increase adaptation time due to their alignmen…

TokenVerse : Towards Flexible Multitask Learning with Dynamic Task Activation
Shashi Kumar, Srikanth Madikeri, Esa\'u Villatoro-Tello, Sergio Burdisso, Pradeep Rangappa, Andr\'es Carofilis, Petr Motlicek, Karthik Pandia, Shankar Venkatesan, Kadri Hacio\u{g}lu, Andreas Stolcke
https://arxiv.org/abs/2508.19856

TokenVerse++: Towards Flexible Multitask Learning with Dynamic Task Activation
Token-based multitasking frameworks like TokenVerse require all training utterances to have labels for all tasks, hindering their ability to leverage partially annotated datasets and scale effectively. We propose TokenVerse++, which introduces learnable vectors in the acoustic embedding space of the XLSR-Transducer ASR model for dynamic task activation. This core mechanism enables training with utterances labeled for only a subset of tasks, a key advantage over TokenVerse. We demonstrate this b…

SupCLAP: Controlling Optimization Trajectory Drift in Audio-Text Contrastive Learning with Support Vector Regularization
Jiehui Luo, Yuguo Yin, Yuxin Xie, Jinghan Ru, Xianwei Zhuang, Minghua He, Aofan Liu, Zihan Xiong, Dongchao Yang
https://arxiv.org/abs/2509.21033

SupCLAP: Controlling Optimization Trajectory Drift in Audio-Text Contrastive Learning with Support Vector Regularization
Contrastive language-audio pretraining, which aims to unify multimodal representations in a shared embedding space, serves as a cornerstone for building a wide range of applications, from cross-modal retrieval to cutting-edge multimodal large language models. However, we find that the perpendicular component of the pushing force from negative samples in contrastive learning is a double-edged sword: it contains rich supplementary information from negative samples, yet its unconstrained nature ca…

Hierarchical Retrieval: The Geometry and a Pretrain-Finetune Recipe
Chong You, Rajesh Jayaram, Ananda Theertha Suresh, Robin Nittka, Felix Yu, Sanjiv Kumar
https://arxiv.org/abs/2509.16411

Hierarchical Retrieval: The Geometry and a Pretrain-Finetune Recipe
Dual encoder (DE) models, where a pair of matching query and document are embedded into similar vector representations, are widely used in information retrieval due to their simplicity and scalability. However, the Euclidean geometry of the embedding space limits the expressive power of DEs, which may compromise their quality. This paper investigates such limitations in the context of hierarchical retrieval (HR), where the document set has a hierarchical structure and the matching documents for…