Tootfinder

Rethinking the Privacy of Text Embeddings: A Reproducibility Study of "Text Embeddings Reveal (Almost) As Much As Text"
Dominykas Seputis, Yongkang Li, Karsten Langerak, Serghei Mihailov
https://arxiv.org/abs/2507.07700

Rethinking the Privacy of Text Embeddings: A Reproducibility Study of "Text Embeddings Reveal (Almost) As Much As Text"
Text embeddings are fundamental to many natural language processing (NLP) tasks, extensively applied in domains such as recommendation systems and information retrieval (IR). Traditionally, transmitting embeddings instead of raw text has been seen as privacy-preserving. However, recent methods such as Vec2Text challenge this assumption by demonstrating that controlled decoding can successfully reconstruct original texts from black-box embeddings. The unexpectedly strong results reported by Vec2…

LGND, which uses vector embeddings to analyze geospatial data and has an enterprise app to query it, raised a $9M seed led by Javelin Venture Partners (Tim De Chant/TechCrunch)
https://techcrunch.com/2025/07/10/lgnd-wants-to-make-chatgpt-for-the-earth/

Exclusive: LGND wants to make ChatGPT for the Earth | TechCrunch
LGND has raised $9 million to convert geographic data into vector embeddings to power AI models that understand the Earth.

This https://arxiv.org/abs/2505.17282 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Attention with Trained Embeddings Provably Selects Important Tokens
Token embeddings play a crucial role in language modeling but, despite this practical relevance, their theoretical understanding remains limited. Our paper addresses the gap by characterizing the structure of embeddings obtained via gradient descent. Specifically, we consider a one-layer softmax attention model with a linear head for binary classification, i.e., $\texttt{Softmax}( p^\top E_X^\top ) E_X v = \frac{ \sum_{i=1}^T \exp(p^\top E_{x_i}) E_{x_i}^\top v}{\sum_{j=1}^T \exp(p^\top E_{x_{j…

Equivariant $K$-theory of cellular toroidal embeddings
Alexis Tchoudjem, V. Uma
https://arxiv.org/abs/2506.07867 https://arxiv.org/pd…

Equivariant $K$-theory of cellular toroidal embeddings
In this article we describe the $G_{comp}\times G_{comp}$-equivariant topological $K$-ring of a {\em cellular} toroidal embedding $\mathbb{X}$ of a complex connected reductive algebraic group $G$. In particular, our results extend the results in \cite{u1} and \cite{u2} on the regular embeddings of $G$, to the equivariant topological $K$-ring of a larger class of (possibly singular) cellular toroidal embeddings. They are also a topological analogue of the results in \cite{gon} on t…

Revealing the Hidden Temporal Structure of HubertSoft Embeddings based on the Russian Phonetic Corpus
Anastasia Ananeva, Anton Tomilov, Marina Volkova
https://arxiv.org/abs/2507.06794

Revealing the Hidden Temporal Structure of HubertSoft Embeddings based on the Russian Phonetic Corpus
Self-supervised learning (SSL) models such as Wav2Vec 2.0 and HuBERT have shown remarkable success in extracting phonetic information from raw audio without labelled data. While prior work has demonstrated that SSL embeddings encode phonetic features at the frame level, it remains unclear whether these models preserve temporal structure, specifically, whether embeddings at phoneme boundaries reflect the identity and order of adjacent phonemes. This study investigates the extent to which boundar…

Bridging Logic and Learning: Decoding Temporal Logic Embeddings via Transformers
Sara Candussio, Gaia Saveri, Gabriele Sarti, Luca Bortolussi
https://arxiv.org/abs/2507.07808

Bridging Logic and Learning: Decoding Temporal Logic Embeddings via Transformers
Continuous representations of logic formulae allow us to integrate symbolic knowledge into data-driven learning algorithms. If such embeddings are semantically consistent, i.e. if similar specifications are mapped into nearby vectors, they enable continuous learning and optimization directly in the semantic space of formulae. However, to translate the optimal continuous representation into a concrete requirement, such embeddings must be invertible. We tackle this issue by training a Transformer…

Extracting Information About Publication Venues Using Citation-Informed Transformers
Brian D. Zimmerman, Joshua Folkins, Olga Vechtomova
https://arxiv.org/abs/2506.08199

Extracting Information About Publication Venues Using Citation-Informed Transformers
Scientific document embeddings contain a variety of rich features which can be harnessed for downstream tasks such as recommendation, ranking, and clustering. We explore which tangible insights can be drawn from scientific document embeddings to understand trends in computer science research featured across nine well-known venues. We collect approximately 60,000 scientific documents published between 2015 and 2023 and analyze their embeddings, which we produce with the SPECTER pre-trained langu…

Platform for Representation and Integration of multimodal Molecular Embeddings
Erika Yilin Zheng, Yu Yan, Baradwaj Simha Sankar, Ethan Ji, Steven Swee, Irsyad Adam, Ding Wang, Alexander Russell Pelletier, Alex Bui, Wei Wang, Peipei Ping
https://arxiv.org/abs/2507.07367

Platform for Representation and Integration of multimodal Molecular Embeddings
Existing machine learning methods for molecular (e.g., gene) embeddings are restricted to specific tasks or data modalities, limiting their effectiveness within narrow domains. As a result, they fail to capture the full breadth of gene functions and interactions across diverse biological contexts. In this study, we have systematically evaluated knowledge representations of biomolecules across multiple dimensions representing a task-agnostic manner spanning three major data sources, including om…

Towards an Explainable Comparison and Alignment of Feature Embeddings
Mohammad Jalali, Bahar Dibaei Nia, Farzan Farnia
https://arxiv.org/abs/2506.06231 htt…

Towards an Explainable Comparison and Alignment of Feature Embeddings
While several feature embedding models have been developed in the literature, comparisons of these embeddings have largely focused on their numerical performance in classification-related downstream applications. However, an interpretable comparison of different embeddings requires identifying and analyzing mismatches between sample groups clustered within the embedding spaces. In this work, we propose the \emph{Spectral Pairwise Embedding Comparison (SPEC)} framework to compare embeddings and …

Piecewise-linear embeddings of the space of 3D lattices into $\RR^{13}$ for high-throughput handling of lattice parameters
Ryoko Oishi-Tomiyasu
https://arxiv.org/abs/2506.08934

Piecewise-linear embeddings of the space of 3D lattices into $\RR^{13}$ for high-throughput handling of lattice parameters
We present two methods to continuously and piecewise-linearly parametrize rank-3 lattices by vectors of $\RR^{13}$, which provides an efficient way to judge if two sets of parameters provide nearly identical lattices within their margins of errors. Such a parametrization can be used to speed up scientific computing involving periodic structures in $\RR^3$ such as crystal structures, which includes database querying, detection of duplicate entries, and structure generation via deep learning tech…

Systolic inequalities on the sphere from symplectic embeddings
Brayan Ferreira
https://arxiv.org/abs/2506.07674 https://arxiv.org/pdf…

Systolic inequalities on the sphere from symplectic embeddings
We use properties of symplectic capacities that were recently defined by Hutchings to obtain upper bounds on the minimal action of Reeb orbits on fiberwise star-shaped hypersurfaces $Σ\subset T^*S^2$. In addition, we introduce the notion of a fiberwise $β$-balanced hypersurface $Σ\subset T^*S^2$ and establish upper bounds for the systole in terms of $β$ and geometric data, in the case of Riemannian metrics on $S^2$ satisfying this property. Finally, under the assumption of antipodal symmetr…

Heterogeneous Sequel-Aware Graph Neural Networks for Sequential Learning
Anushka Tiwari, Haimonti Dutta, Shahrzad Khanizadeh
https://arxiv.org/abs/2506.05625

Heterogeneous Sequel-Aware Graph Neural Networks for Sequential Learning
Graph-based recommendation systems use higher-order user and item embeddings for next-item predictions. Dynamically adding collaborative signals from neighbors helps to use similar users' preferences during learning. While item-item correlations and their impact on recommendations have been studied, the efficacy of temporal item sequences for recommendations is much less explored. In this paper, we examine temporal item sequence (sequel-aware) embeddings along with higher-order user embeddings …

Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection
Subhajit Maity, Ayan Kumar Bhunia, Subhadeep Koley, Pinaki Nath Chowdhury, Aneeshan Sain, Yi-Zhe Song
https://arxiv.org/abs/2507.07994

Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection
Keypoint detection, integral to modern machine perception, faces challenges in few-shot learning, particularly when source data from the same distribution as the query is unavailable. This gap is addressed by leveraging sketches, a popular form of human expression, providing a source-free alternative. However, challenges arise in mastering cross-modal embeddings and handling user-specific sketch styles. Our proposed framework overcomes these hurdles with a prototypical setup, combined with a gr…

Embeddings of Sobolev spaces of differential forms
Vladimir Gol'dshtein, Yaroslav Kopylov, Roman Panenko
https://arxiv.org/abs/2507.05851 https://

Embeddings of Sobolev spaces of differential forms
We continue the~study of embedding between different classes of Sobolev spaces of differential forms started in 2006 in a~paper by Gol'dshtein and Troyanov. As in this paper, our study is based on relations between $L_{q,p}$-cohomology and Sobolev type inequalities. The~main results are estimates for the norms of the embedding operators for $q=p$ and $p>\frac{n-1}{k-1}$ in the~Euclidean $r$-ball $B(r)$ and its bi-Lipschitz images. The~compactness of such operators also studied.

Training-Free Query Optimization via LLM-Based Plan Similarity
Nikita Vasilenko, Alexander Demin, Vladimir Boorlakov
https://arxiv.org/abs/2506.05853 https…

Training-Free Query Optimization via LLM-Based Plan Similarity
Large language model (LLM) embeddings offer a promising new avenue for database query optimization. In this paper, we explore how pre-trained execution plan embeddings can guide SQL query execution without the need for additional model training. We introduce LLM-PM (LLM-based Plan Mapping), a framework that embeds the default execution plan of a query, finds its k nearest neighbors among previously executed plans, and recommends database hintsets based on neighborhood voting. A lightweight cons…

Assessing the Alignment of Audio Representations with Timbre Similarity Ratings
Haokun Tian, Stefan Lattner, Charalampos Saitis
https://arxiv.org/abs/2507.07764

Assessing the Alignment of Audio Representations with Timbre Similarity Ratings
Psychoacoustical so-called "timbre spaces" map perceptual similarity ratings of instrument sounds onto low-dimensional embeddings via multidimensional scaling, but suffer from scalability issues and are incapable of generalization. Recent results from audio (music and speech) quality assessment as well as image similarity have shown that deep learning is able to produce embeddings that align well with human perception while being largely free from these constraints. Although the existing human-…

Efficient and Adaptive Estimation of Local Triadic Coefficients
Ilie Sarpe, Aristides Gionis
https://arxiv.org/abs/2507.07536 https://

Efficient and Adaptive Estimation of Local Triadic Coefficients
Characterizing graph properties is fundamental to the analysis and to our understanding of real-world networked systems. The local clustering coefficient, and the more recently introduced, local closure coefficient, capture powerful properties that are essential in a large number of applications, ranging from graph embeddings to graph partitioning. Such coefficients capture the local density of the neighborhood of each node, considering incident triadic structures and paths of length two. For t…

Neighborhood Overlap-Aware High-Order Graph Neural Network for Dynamic Graph Learning
Ling Wang
https://arxiv.org/abs/2506.06728 https://

Neighborhood Overlap-Aware High-Order Graph Neural Network for Dynamic Graph Learning
Dynamic graph learning (DGL) aims to learn informative and temporally-evolving node embeddings to support downstream tasks such as link prediction. A fundamental challenge in DGL lies in effectively modeling both the temporal dynamics and structural dependencies of evolving graph topologies. Recent advances in Dynamic Graph Neural Networks (DGNNs) have obtained remarkable success by leveraging message-passing mechanisms to capture pairwise node interactions. However, these approaches often over…

Perfect t-embeddings of doubly periodic Aztec diamonds
Tomas Berggren, Matthew Nicoletti, Marianna Russkikh
https://arxiv.org/abs/2508.04938 https://arxiv.…

Perfect t-embeddings of doubly periodic Aztec diamonds
We study the large-scale geometry of t-surfaces -- pairs of perfect t-embeddings and their associated origami maps -- arising from dimer models on Aztec diamonds with periodic edge weights. We prove that these t-surfaces converge to space-like maximal surfaces in the Minkowski space $\mathbb{R}^{2,2}$. We observe that the frozen and gas regions influence the geometry of the limiting surface in striking ways: all frozen regions collapse to four boundary points, regardless of the number of frozen…

FuDoBa: Fusing Document and Knowledge Graph-based Representations with Bayesian Optimisation
Boshko Koloski, Senja Pollak, Roberto Navigli, Bla\v{z} \v{S}krlj
https://arxiv.org/abs/2507.06622

FuDoBa: Fusing Document and Knowledge Graph-based Representations with Bayesian Optimisation
Building on the success of Large Language Models (LLMs), LLM-based representations have dominated the document representation landscape, achieving great performance on the document embedding benchmarks. However, the high-dimensional, computationally expensive embeddings from LLMs tend to be either too generic or inefficient for domain-specific applications. To address these limitations, we introduce FuDoBa a Bayesian optimisation-based method that integrates LLM-based embeddings with domain-spe…

Classification of Equivariant Legendrian Embeddings of Rational Homogeneous Spaces into Nilpotent Orbits
Minseong Kwon
https://arxiv.org/abs/2507.03932 htt…

Classification of Equivariant Legendrian Embeddings of Rational Homogeneous Spaces into Nilpotent Orbits
For a complex semi-simple Lie algebra, every nilpotent orbit in its projectivization comes with a complex contact structure. For each nilpotent orbit, we classify projective Legendrian subvarieties that are homogeneous under the actions of their stabilizers in the adjoint group. In particular, we present a classification of equivariant Legendrian embeddings of rational homogeneous spaces into adjoint varieties.

CaliciBoost: Performance-Driven Evaluation of Molecular Representations for Caco-2 Permeability Prediction
Huong Van Le, Weibin Ren, Junhong Kim, Yukyung Yun, Young Bin Park, Young Jun Kim, Bok Kyung Han, Inho Choi, Jong IL Park, Hwi-Yeol Yun, Jae-Mun Choi
https://arxiv.org/abs/2506.08059

CaliciBoost: Performance-Driven Evaluation of Molecular Representations for Caco-2 Permeability Prediction
Caco-2 permeability serves as a critical in vitro indicator for predicting the oral absorption of drug candidates during early-stage drug discovery. To enhance the accuracy and efficiency of computational predictions, we systematically investigated the impact of eight molecular feature representation types including 2D/3D descriptors, structural fingerprints, and deep learning-based embeddings combined with automated machine learning techniques to predict Caco-2 permeability. Using two datasets…

Universal Embeddings of Tabular Data
Astrid Franz, Frederik Hoppe, Marianne Michaelis, Udo G\"obel
https://arxiv.org/abs/2507.05904 https://

Universal Embeddings of Tabular Data
Tabular data in relational databases represents a significant portion of industrial data. Hence, analyzing and interpreting tabular data is of utmost importance. Application tasks on tabular data are manifold and are often not specified when setting up an industrial database. To address this, we present a novel framework for generating universal, i.e., task-independent embeddings of tabular data for performing downstream tasks without predefined targets. Our method transforms tabular data into …

More Exotic $\mathbb{RP}^2$-knots and Homotopy Spheres
Judson Kuhrman
https://arxiv.org/abs/2507.03798 https://arxiv.org/pdf/2507.037…

More Exotic $\mathbb{RP}^2$-knots and Homotopy Spheres
We extend the infinite family of exotic embeddings $\mathbb{RP}^2 \hookrightarrow S^4$ constructed by Miyazawa to a strictly larger family of exotic embeddings, by showing that in place of the pretzel knot $P(-2, 3, 7)$, an infinite family of knots may be used as input to the construction. To this end, we prove that for any Montesinos knot of the form $K(2,3,|6s+1|)$, the branched double cover of the corresponding roll-spun knot is a homotopy sphere. This in turn produces a larger fam…

This https://arxiv.org/abs/2409.04459 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

WET: Overcoming Paraphrasing Vulnerabilities in Embeddings-as-a-Service with Linear Transformation Watermarks
Embeddings-as-a-Service (EaaS) is a service offered by large language model (LLM) developers to supply embeddings generated by LLMs. Previous research suggests that EaaS is prone to imitation attacks -- attacks that clone the underlying EaaS model by training another model on the queried embeddings. As a result, EaaS watermarks are introduced to protect the intellectual property of EaaS providers. In this paper, we first show that existing EaaS watermarks can be removed by paraphrasing when att…

Dimension-Preserving Saturated Embeddings of Finite Posets into the Spectra of Noetherian UFDs
David Baron, S. Loepp
https://arxiv.org/abs/2507.03574 https…

Dimension-Preserving Saturated Embeddings of Finite Posets into the Spectra of Noetherian UFDs
Given a finite poset $X$, we find necessary and sufficient conditions for there to exist a local Noetherian UFD $A$ and a saturated embedding of posets $ϕ: X \longrightarrow \mbox{Spec}(A)$ such that $\dim(X)=\dim(A)$. The conditions imposed on $X$ in our characterization are remarkably mild, demonstrating that there is a large class of finite posets that can be embedded into the spectrum of a local Noetherian UFD of the same dimension as $X$ in a way that preserves saturated chains. We also s…

A free two-generated left distributive algebra of elementary embeddings
Andrew D. Brooke-Taylor, Scott Cramer, Sheila K. Miller Edwards
https://arxiv.org/abs/2508.02244 https://…

A free two-generated left distributive algebra of elementary embeddings
The set-theoretic large cardinal axiom known as I3 posits the existence of a non-trivial rank-to-rank embedding from an initial segment of the universe of sets into itself. Laver showed that the algebra generated by a single such embedding under the operation of application is in fact the free left distributive (LD) algebra on one generator. This and associated theorems using the set-theoretic structure of the embeddings yielded numerous results about general LD algebras under the assumption of…

jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval
Michael G\"unther, Saba Sturua, Mohammad Kalim Akram, Isabelle Mohr, Andrei Ungureanu, Sedigheh Eslami, Scott Martens, Bo Wang, Nan Wang, Han Xiao
https://arxiv.org/abs/2506.18902

jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval
We introduce jina-embeddings-v4, a 3.8 billion parameter multimodal embedding model that unifies text and image representations through a novel architecture supporting both single-vector and multi-vector embeddings in the late interaction style. The model incorporates task-specific Low-Rank Adaptation (LoRA) adapters to optimize performance across diverse retrieval scenarios, including query-based information retrieval, cross-modal semantic similarity, and programming code search. Comprehensive…

This https://arxiv.org/abs/2502.19311 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLO_…

Faithful Logic Embeddings in HOL -- Deep and Shallow
Deep and shallow embeddings of non-classical logics in classical higher-order logic have been explored, implemented, and used in various reasoning tools in recent years. This paper presents a method for the simultaneous deployment of deep and shallow embeddings of various degrees in classical higher-order logic. This enables flexible, interactive and automated theorem proving and counterexample finding at meta and object level, as well as automated faithfulness proofs between these logic embedd…

Do We Really Need Specialization? Evaluating Generalist Text Embeddings for Zero-Shot Recommendation and Search
Matteo Attimonelli, Alessandro De Bellis, Claudio Pomo, Dietmar Jannach, Eugenio Di Sciascio, Tommaso Di Noia
https://arxiv.org/abs/2507.05006

Do We Really Need Specialization? Evaluating Generalist Text Embeddings for Zero-Shot Recommendation and Search
Pre-trained language models (PLMs) are widely used to derive semantic representations from item metadata in recommendation and search. In sequential recommendation, PLMs enhance ID-based embeddings through textual metadata, while in product search, they align item characteristics with user intent. Recent studies suggest task and domain-specific fine-tuning are needed to improve representational power. This paper challenges this assumption, showing that Generalist Text Embedding Models (GTEs), p…

Reproducibility in embedding benchmarks is challenging, especially with embedding models that are instructional and increasingly large. At #bbuzz, Isaac Chung explained how MTEB addresses prompt variability, scaling issues and large datasets to ensure fair and consistent scores, setting a standard for benchmarking embeddings.
Watch the full session:

This https://arxiv.org/abs/2505.12156 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

Quiver Schemes for Nongeneric Stability and Cornering
We use the Le Bruyn--Procesi theorem to prove several results on quiver schemes for nongeneric stability conditions: We show that the stability-zero finite-type Nakajima quiver schemes are reduced points and give an example of a closely related nonreduced quiver scheme. In broader generality, we prove that adding modules supported on stability-zero vertices induces closed embeddings of quiver schemes, and show how in many cases the quiver scheme associated with the cornered algebra defines a li…

This https://arxiv.org/abs/2506.04678 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDB_…

BVLSM: Write-Efficient LSM-Tree Storage via WAL-Time Key-Value Separation
Modern data-intensive applications increasingly store and process big-value items, such as multimedia objects and machine learning embeddings, which exacerbate storage inefficiencies in Log-Structured Merge-Tree (LSM)-based key-value stores. This paper presents BVLSM, a Write-Ahead Log (WAL)-time key-value separation mechanism designed to address three key challenges in LSM-Tree storage systems: write amplification, poor memory utilization, and I/O jitter under big-value workloads. Unlike state…

Emergent Semantics Beyond Token Embeddings: Transformer LMs with Frozen Visual Unicode Representations
A. Bochkov
https://arxiv.org/abs/2507.04886 https://…

Emergent Semantics Beyond Token Embeddings: Transformer LMs with Frozen Visual Unicode Representations
Understanding the locus of semantic representation in large language models (LLMs) is crucial for interpretability and architectural innovation. The dominant paradigm posits that trainable input embeddings serve as foundational "meaning vectors." This paper challenges that view. We construct Transformer models where the embedding layer is entirely frozen, with vectors derived not from data, but from the visual structure of Unicode glyphs. These non-semantic, precomputed visual embeddings are fi…

CSI2Vec: Towards a Universal CSI Feature Representation for Positioning and Channel Charting
Victoria Palhares, Sueda Taner, Christoph Studer
https://arxiv.org/abs/2506.05237

CSI2Vec: Towards a Universal CSI Feature Representation for Positioning and Channel Charting
Natural language processing techniques, such as Word2Vec, have demonstrated exceptional capabilities in capturing semantic and syntactic relationships of text through vector embeddings. Inspired by this technique, we propose CSI2Vec, a self-supervised framework for generating universal and robust channel state information (CSI) representations tailored to CSI-based positioning (POS) and channel charting (CC). CSI2Vec learns compact vector embeddings across various wireless scenarios, capturing …

Charge quantisation, monopoles and emergent symmetry in the Standard Model and its embeddings
Rodrigo Alonso, Despoina Dimakou, Yunji Ha, Valentin V. Khoze
https://arxiv.org/abs/2507.01777

Charge quantisation, monopoles and emergent symmetry in the Standard Model and its embeddings
This work studies the connection of the global properties of the SM gauge group to 1-form discrete symmetries, the possible non-Abelian embeddings of the SM group, and electric and magnetic charge quantisation. Building on previous work, we introduce indexes to characterise the group choices, connect the concept of compositeness degree to emergent electric 1-form symmetry, introduce a new model to fill in the $p=1$ gap, and analyse the magnetic spectrum while connecting its UV and IR realisatio…

Replaced article(s) found for math.SG. https://arxiv.org/list/math.SG/new
[1/1]:
- Rigid-flexible values for symplectic embeddings of four-dimensional ellipsoids into almost-cubes
Andrew Lee, Cory H. Colbert

Federated Learning for ICD Classification with Lightweight Models and Pretrained Embeddings
Binbin Xu, G\'erard Dray
https://arxiv.org/abs/2507.03122 h…

Federated Learning for ICD Classification with Lightweight Models and Pretrained Embeddings
This study investigates the feasibility and performance of federated learning (FL) for multi-label ICD code classification using clinical notes from the MIMIC-IV dataset. Unlike previous approaches that rely on centralized training or fine-tuned large language models, we propose a lightweight and scalable pipeline combining frozen text embeddings with simple multilayer perceptron (MLP) classifiers. This design offers a privacy-preserving and deployment-efficient alternative for clinical NLP app…

Embeddings of Weighted Morrey Spaces
Marcus Gerhold
https://arxiv.org/abs/2508.03387 https://arxiv.org/pdf/2508.03387…

Embeddings of Weighted Morrey Spaces
In this master thesis we recall already established definitions and basic properties of classical Morrey spaces in an attempt to expand known facts to their weighted counterparts. To do so, we will recall properties of Muckenhoupt weights, which we will use to derive further properties of weighted Morrey spaces. We will also show the boundedness of the Hardy-Littlewood maximal operator on weighted Morrey spaces. Throughout this thesis we will have a look at example weights and show in which cas…

Factorizable embeddings and the period of an irreducible sofic shift
Brian Marcus, Tom Meyerovitch, Klaus Thomsen, Chengyu Wu
https://arxiv.org/abs/2508.02554 https://

Factorizable embeddings and the period of an irreducible sofic shift
Generalizing a result of MacDonald we give necessary and sufficient conditions for an arbitrary subshift to embed into an irreducible sofic shift factoring through a given cover by an irreducible subshift of finite type (SFT). We obtain also necessary and sufficient conditions for an arbitrary subshift to embed into an irreducible sofic shift factoring through \emph{some} sliding block code out of an irreducible SFT. We do that when the code is required to be surjective, and hence a factor code…

Text adaptation for speaker verification with speaker-text factorized embeddings
Yexin Yang, Shuai Wang, Xun Gong, Yanmin Qian, Kai Yu
https://arxiv.org/abs/2508.04425 https://

Text adaptation for speaker verification with speaker-text factorized embeddings
Text mismatch between pre-collected data, either training data or enrollment data, and the actual test data can significantly hurt text-dependent speaker verification (SV) system performance. Although this problem can be solved by carefully collecting data with the target speech content, such data collection could be costly and inflexible. In this paper, we propose a novel text adaptation framework to address the text mismatch issue. Here, a speaker-text factorization network is proposed to fac…

Robust Target Speaker Diarization and Separation via Augmented Speaker Embedding Sampling
Md Asif Jalal, Luca Remaggi, Vasileios Moschopoulos, Thanasis Kotsiopoulos, Vandana Rajan, Karthikeyan Saravanan, Anastasis Drosou, Junho Heo, Hyuk Oh, Seokyeong Jeong
https://arxiv.org/abs/2508.06393

Robust Target Speaker Diarization and Separation via Augmented Speaker Embedding Sampling
Traditional speech separation and speaker diarization approaches rely on prior knowledge of target speakers or a predetermined number of participants in audio signals. To address these limitations, recent advances focus on developing enrollment-free methods capable of identifying targets without explicit speaker labeling. This work introduces a new approach to train simultaneous speech separation and diarization using automatic identification of target speaker embeddings, within mixtures. Our p…

Measuring Information Richness in Product Images: Implications for Online Sales
Zhu Yuting, Cao Xinyu, Su Yuzhuo, Ma Yongbin
https://arxiv.org/abs/2508.04541 https://

Measuring Information Richness in Product Images: Implications for Online Sales
A common challenge for e-commerce sellers is to decide what product images to display on online shopping sites. In this paper, we propose and validate a novel metric, k-value, to quantify the information richness of an image set, and we further investigate its effect on consumers' purchase decisions. We leverage patch-level embeddings from Vision Transformers (ViT) and apply k-means clustering to identify distinct visual features, defining k-value as the number of clusters. An online experiment…

Associative triple trisystems and standard embeddings
Ra\'ul Felipe, Guillermo Vera de Salas
https://arxiv.org/abs/2506.04191 https://

Associative triple trisystems and standard embeddings
Building on the established theories of Jordan triple disystems and Leibniz triple systems, we introduce and develop the theory of associative triple trisystems, filling a significant gap in the existing framework. We establish the classical relationships between associative, Jordan, and Lie triple systems within the context of trisystems. We present a significant example by equipping the space of matrices with a non-trivial associative dialgebra structure. We conclude defining the concept of d…

Learning to cluster neuronal function
Nina S. Nellen, Polina Turishcheva, Michaela Vystr\v{c}ilov\'a, Shashwat Sridhar, Tim Gollisch, Andreas S. Tolias, Alexander S. Ecker
https://arxiv.org/abs/2506.03293

Learning to cluster neuronal function
Deep neural networks trained to predict neural activity from visual input and behaviour have shown great potential to serve as digital twins of the visual cortex. Per-neuron embeddings derived from these models could potentially be used to map the functional landscape or identify cell types. However, state-of-the-art predictive models of mouse V1 do not generate functional embeddings that exhibit clear clustering patterns which would correspond to cell types. This raises the question whether th…

A genuine equivariant recognition principle for finite groups
Branko Juran
https://arxiv.org/abs/2508.04421 https://arxiv.org/pdf/2508.04421

A genuine equivariant recognition principle for finite groups
For $G$ a finite group and $V$ a finite dimensional real $G$-representation, there is a $G$-operad $\mathbb{E}_{V}$ defined using embeddings of $V$-framed $G$-disks such that for any based $G$-space $X$, there is a naturally defined $\mathbb{E}_{V}$-algebra structure on the $V$-fold space $Ω^V X$. Given an $\mathbb{E}_{V}$-algebra in $G$-spaces and a subgroup $H$ of $G$, the fixed points $A^H$ carry the structure of an $\mathbb{E}_{\dim V^H}$-algebra in spaces. We prove that an $\mathbb{E}_{V}…

Infinity Search: Approximate Vector Search with Projections on q-Metric Spaces
Antonio Pariente, Ignacio Hounie, Santiago Segarra, Alejandro Ribeiro
https://arxiv.org/abs/2506.06557

Infinity Search: Approximate Vector Search with Projections on q-Metric Spaces
Despite the ubiquity of vector search applications, prevailing search algorithms overlook the metric structure of vector embeddings, treating it as a constraint rather than exploiting its underlying properties. In this paper, we demonstrate that in $q$-metric spaces, metric trees can leverage a stronger version of the triangle inequality to reduce comparisons for exact search. Notably, as $q$ approaches infinity, the search complexity becomes logarithmic. Therefore, we propose a novel projectio…

Evaluating Style-Personalized Text Generation: Challenges and Directions
Anubhav Jangra, Bahareh Sarrafzadeh, Adrian de Wynter, Silviu Cucerzan, Sujay Kumar Jauhar
https://arxiv.org/abs/2508.06374

Evaluating Style-Personalized Text Generation: Challenges and Directions
While prior research has built tools and benchmarks towards style personalized text generation, there has been limited exploration of evaluation in low-resource author style personalized text generation space. Through this work, we question the effectiveness of the widely adopted evaluation metrics like BLEU and ROUGE, and explore other evaluation paradigms such as style embeddings and LLM-as-judge to holistically evaluate the style personalized text generation task. We evaluate these metrics a…

Beyond Distance: Mobility Neural Embeddings Reveal Visible and Invisible Barriers in Urban Space
Guangyuan Weng, Minsuk Kim, Yong-Yeol Ahn, Esteban Moro
https://arxiv.org/abs/2506.24061

Beyond Distance: Mobility Neural Embeddings Reveal Visible and Invisible Barriers in Urban Space
Human mobility in cities is shaped not only by visible structures such as highways, rivers, and parks but also by invisible barriers rooted in socioeconomic segregation, uneven access to amenities, and administrative divisions. Yet identifying and quantifying these barriers at scale and their relative importance on people's movements remains a major challenge. Neural embedding models, originally developed for language, offer a powerful way to capture the complexity of human mobility from large-…

Conformal Rigidity and Spectral Embeddings of Graphs
Jo\~ao Gouveia, Stefan Steinerberger, Rekha R. Thomas
https://arxiv.org/abs/2506.20541 https://…

Conformal Rigidity and Spectral Embeddings of Graphs
We investigate the structure of conformally rigid graphs. Graphs are conformally rigid if introducing edge weights cannot increase (decrease) the second (last) eigenvalue of the Graph Laplacian. Edge-transitive graphs and distance-regular graphs are known to be conformally rigid. We establish new results using the connection between conformal rigidity and edge-isometric spectral embeddings of the graph. All $1$-walk regular graphs are conformally rigid, a consequence of a stronger property of t…

Uncovering smooth structures in single-cell data with PCS-guided neighbor embeddings
Rong Ma, Xi Li, Jingyuan Hu, Bin Yu
https://arxiv.org/abs/2506.22228 h…

Uncovering smooth structures in single-cell data with PCS-guided neighbor embeddings
Single-cell sequencing is revolutionizing biology by enabling detailed investigations of cell-state transitions. Many biological processes unfold along continuous trajectories, yet it remains challenging to extract smooth, low-dimensional representations from inherently noisy, high-dimensional single-cell data. Neighbor embedding (NE) algorithms, such as t-SNE and UMAP, are widely used to embed high-dimensional single-cell data into low dimensions. But they often introduce undesirable distortio…

Embedded contact homology of the unit cotangent bundle of the Klein bottle
Marcelo Miranda, Vinicius G. B. Ramos
https://arxiv.org/abs/2508.06400 https://a…

Embedded contact homology of the unit cotangent bundle of the Klein bottle
We give a combinatorial description of the embedded contact homology chain complex of the unit cotangent bundle of the Klein bottle with the standard flat Riemannian metric. Using pseudoholomorphic curves coming from the associated differential, we find an obstruction theorem for symplectic embeddings of toric domains $X_Ω\subset \mathbb{C}^2$ into the unit disk cotangent bundle $D^*K$. As an application we compute the Gromov width of $D^*K$.

A Study of the Framework and Real-World Applications of Language Embedding for 3D Scene Understanding
Mahmoud Chick Zaouali, Todd Charter, Yehor Karpichev, Brandon Haworth, Homayoun Najjjaran
https://arxiv.org/abs/2508.05064

A Study of the Framework and Real-World Applications of Language Embedding for 3D Scene Understanding
Gaussian Splatting has rapidly emerged as a transformative technique for real-time 3D scene representation, offering a highly efficient and expressive alternative to Neural Radiance Fields (NeRF). Its ability to render complex scenes with high fidelity has enabled progress across domains such as scene reconstruction, robotics, and interactive content creation. More recently, the integration of Large Language Models (LLMs) and language embeddings into Gaussian Splatting pipelines has opened new …

News Sentiment Embeddings for Stock Price Forecasting
Ayaan Qayyum
https://arxiv.org/abs/2507.01970 https://arxiv.org/pdf/2507.01970

News Sentiment Embeddings for Stock Price Forecasting
This paper will discuss how headline data can be used to predict stock prices. The stock price in question is the SPDR S&P 500 ETF Trust, also known as SPY that tracks the performance of the largest 500 publicly traded corporations in the United States. A key focus is to use news headlines from the Wall Street Journal (WSJ) to predict the movement of stock prices on a daily timescale with OpenAI-based text embedding models used to create vector encodings of each headline with principal componen…

Quasi-isometric embeddings of Ramanujan complexes
Hyein Choi
https://arxiv.org/abs/2506.23585 https://arxiv.org/pdf/2506.23585…

Quasi-isometric embeddings of Ramanujan complexes
Ramanujan complexes were defined as high dimensional analogues of the optimal expanders, Ramanujan graphs. They were constructed as quotients of the Euclidean building (also called the affine building and the Bruhat-Tits building) of $\mathrm{PGL}_d(\mathbb{F}_p((y)))$ for any prime $p$ by Lubotzky-Samuels-Vishne. We distinguish the Ramanujan complexes up to large-scale geometry. More precisely, we show that if $p$ and $q$ are distinct primes, then the associated Ramanujan complexes do not quas…

Compressing Large Language Models with PCA Without Performance Loss
Magnus Bengtsson
https://arxiv.org/abs/2508.04307 https://arxiv.org/pdf/2508.04307

Compressing Large Language Models with PCA Without Performance Loss
We demonstrate that Principal Component Analysis (PCA), when applied in a structured manner, either to polar-transformed images or segment-wise to token sequences, enables extreme compression of neural models without sacrificing performance. Across three case studies, we show that a one-layer classifier trained on PCA-compressed polar MNIST achieves over 98 percent accuracy using only 840 parameters. A two-layer transformer trained on 70-dimensional PCA-reduced MiniLM embeddings reaches 76.62 p…

One of our final topics in the #ISE2025 lecture were Knowledge Graph Embeddings. How to vectorise KG structures while preserving their inherent semantics?
#AI #KGE

KAConvText: Novel Approach to Burmese Sentence Classification using Kolmogorov-Arnold Convolution
Ye Kyaw Thu, Thura Aung, Thazin Myint Oo, Thepchai Supnithi
https://arxiv.org/abs/2507.06753

KAConvText: Novel Approach to Burmese Sentence Classification using Kolmogorov-Arnold Convolution
This paper presents the first application of Kolmogorov-Arnold Convolution for Text (KAConvText) in sentence classification, addressing three tasks: imbalanced binary hate speech detection, balanced multiclass news classification, and imbalanced multiclass ethnic language identification. We investigate various embedding configurations, comparing random to fastText embeddings in both static and fine-tuned settings, with embedding dimensions of 100 and 300 using CBOW and Skip-gram models. Baselin…

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…

Toroidal embedding of Chevalley groups over $\mathbb{Z}$
Shang Li
https://arxiv.org/abs/2506.02638 https://arxiv.org/pdf/2506.02638…

Toroidal embedding of Chevalley groups over $\mathbb{Z}$
The classification of equivariant toroidal embeddings of a reductive group over an algebraically closed field is combinatorial and does not depend on the characteristic of the base field. This suggests that there should exist ``universal'' toroidal embeddings for a Chevalley group scheme over $\mathbb{Z}$ which specialize to classical toroidal embeddings via base change. In this paper, we establish the existence of ``universal'' equivariant toroidal embeddings for split reductive group schemes …

Robust Learning on Noisy Graphs via Latent Space Constraints with External Knowledge
Chunhui Gu, Mohammad Sadegh Nasr, James P. Long, Kim-Anh Do, Ehsan Irajizad
https://arxiv.org/abs/2507.05540

Robust Learning on Noisy Graphs via Latent Space Constraints with External Knowledge
Graph Neural Networks (GNNs) often struggle with noisy edges. We propose Latent Space Constrained Graph Neural Networks (LSC-GNN) to incorporate external "clean" links and guide embeddings of a noisy target graph. We train two encoders--one on the full graph (target plus external edges) and another on a regularization graph excluding the target's potentially noisy links--then penalize discrepancies between their latent representations. This constraint steers the model away from overfitting spur…

Towards Storage-Efficient Visual Document Retrieval: An Empirical Study on Reducing Patch-Level Embeddings
Yubo Ma, Jinsong Li, Yuhang Zang, Xiaobao Wu, Xiaoyi Dong, Pan Zhang, Yuhang Cao, Haodong Duan, Jiaqi Wang, Yixin Cao, Aixin Sun
https://arxiv.org/abs/2506.04997

Towards Storage-Efficient Visual Document Retrieval: An Empirical Study on Reducing Patch-Level Embeddings
Despite the strong performance of ColPali/ColQwen2 in Visualized Document Retrieval (VDR), it encodes each page into multiple patch-level embeddings and leads to excessive memory usage. This empirical study investigates methods to reduce patch embeddings per page at minimum performance degradation. We evaluate two token-reduction strategies: token pruning and token merging. Regarding token pruning, we surprisingly observe that a simple random strategy outperforms other sophisticated pruning met…

This https://arxiv.org/abs/2502.19311 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLO_…

Faithful Logic Embeddings in HOL -- Deep and Shallow
Deep and shallow embeddings of non-classical logics in classical higher-order logic have been explored, implemented, and used in various reasoning tools in recent years. This paper presents a method for the simultaneous deployment of deep and shallow embeddings of various degrees in classical higher-order logic. This enables flexible, interactive and automated theorem proving and counterexample finding at meta and object level, as well as automated faithfulness proofs between these logic embedd…

Synopsis: Secure and private trend inference from encrypted semantic embeddings
Madelyne Xiao, Palak Jain, Micha Gorelick, Sarah Scheffler
https://arxiv.org/abs/2505.23880

Synopsis: Secure and private trend inference from encrypted semantic embeddings
WhatsApp and many other commonly used communication platforms guarantee end-to-end encryption (E2EE), which requires that service providers lack the cryptographic keys to read communications on their own platforms. WhatsApp's privacy-preserving design makes it difficult to study important phenomena like the spread of misinformation or political messaging, as users have a clear expectation and desire for privacy and little incentive to forfeit that privacy in the process of handing over raw data…

Advancement of Circular Economy Through Interdisciplinary Collaboration: A Bibliometric Approach
Keita Nishimoto, Koji Kimita, Shinsuke Murakami, Yin Long, Kimitaka Asatani, Ichiro Sakata
https://arxiv.org/abs/2507.04923

Advancement of Circular Economy Through Interdisciplinary Collaboration: A Bibliometric Approach
Since the European Union introduced its Circular Economy (CE) Action Plan in 2015, CE research has expanded rapidly. However, the structure of this emerging field - both in terms of its constituent disciplines and researcher dynamics - remains poorly understood. To address this gap, we analyze over 25,000 CE-related publications from Scopus by combining conventional bibliometric approaches with advanced machine learning techniques, including text embeddings and clustering. This hybrid method en…

This https://arxiv.org/abs/2502.04049 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Towards Explainable Spoofed Speech Attribution and Detection:a Probabilistic Approach for Characterizing Speech Synthesizer Components
We propose an explainable probabilistic framework for characterizing spoofed speech by decomposing it into probabilistic attribute embeddings. Unlike raw high-dimensional countermeasure embeddings, which lack interpretability, the proposed probabilistic attribute embeddings aim to detect specific speech synthesizer components, represented through high-level attributes and their corresponding values. We use these probabilistic embeddings with four classifier back-ends to address two downstream t…

This https://arxiv.org/abs/2505.14806 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_qbi…

Place Cells as Position Embeddings of Multi-Time Random Walk Transition Kernels for Path Planning
The hippocampus orchestrates spatial navigation through collective place cell encodings that form cognitive maps. We reconceptualize the population of place cells as position embeddings approximating multi-scale symmetric random walk transition kernels: the inner product $\langle h(x, t), h(y, t) \rangle = q(y|x, t)$ represents normalized transition probabilities, where $h(x, t)$ is the embedding at location $ x $, and $q(y|x, t)$ is the normalized symmetric transition probability ove…

Cropping outperforms dropout as an augmentation strategy for training self-supervised text embeddings
Rita Gonz\'alez-M\'arquez, Philipp Berens, Dmitry Kobak
https://arxiv.org/abs/2508.03453

Cropping outperforms dropout as an augmentation strategy for training self-supervised text embeddings
Text embeddings, i.e. vector representations of entire texts, play an important role in many NLP applications, such as retrieval-augmented generation, sentiment analysis, clustering, or visualizing collections of texts for data exploration. Currently, top-performing embedding models are derived from pre-trained language models via extensive supervised fine-tuning using curated text pairs. This contrasts with computer vision, where self-supervised training based on data augmentations has demonst…

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[8/10]:
- Preserving Topological and Geometric Embeddings for Point Cloud Recovery
Kaiyue Zhou, Zelong Tan, Hongxiao Wang, Ya-li Li, Shengjin Wang

WAKE: Watermarking Audio with Key Enrichment
Yaoxun Xu, Jianwei Yu, Hangting Chen, Zhiyong Wu, Xixin Wu, Dong Yu, Rongzhi Gu, Yi Luo
https://arxiv.org/abs/2506.05891

WAKE: Watermarking Audio with Key Enrichment
As deep learning advances in audio generation, challenges in audio security and copyright protection highlight the need for robust audio watermarking. Recent neural network-based methods have made progress but still face three main issues: preventing unauthorized access, decoding initial watermarks after multiple embeddings, and embedding varying lengths of watermarks. To address these issues, we propose WAKE, the first key-controllable audio watermark framework. WAKE embeds watermarks using sp…

Multiplier Between Generalized Toeplitz Kernels
Anjali, R. K. Srivastava
https://arxiv.org/abs/2507.03452 https://arxiv.org/pdf/2507.…

Multiplier Between Generalized Toeplitz Kernels
We develop a structural classification of multipliers between generalized Toeplitz kernels, extending work of Fricain and Rishika. Our results establish new equivalences between multiplier algebras and Carleson-type embeddings, linking them to Beurling Malliavin densities, Pólya sequences, and the spectral theory of entire functions.

Twisted embeddings of tori have small extrinsic systole
Sahana Vasudevan
https://arxiv.org/abs/2507.23766 https://arxiv.org/pdf/2507.23766

Twisted embeddings of tori have small extrinsic systole
We prove a type of systolic inequality for embeddings of $T^2$ in $\mathbb{R}^3$. In particular, a highly twisted $T^2$ embedded in $\mathbb{R}^3$ must contain a non-contractible loop of small $\mathbb{R}^3$-diameter.

EAGLE: Efficient Alignment of Generalized Latent Embeddings for Multimodal Survival Prediction with Interpretable Attribution Analysis
Aakash Tripathi, Asim Waqas, Matthew B. Schabath, Yasin Yilmaz, Ghulam Rasool
https://arxiv.org/abs/2506.22446

EAGLE: Efficient Alignment of Generalized Latent Embeddings for Multimodal Survival Prediction with Interpretable Attribution Analysis
Accurate cancer survival prediction requires integration of diverse data modalities that reflect the complex interplay between imaging, clinical parameters, and textual reports. However, existing multimodal approaches suffer from simplistic fusion strategies, massive computational requirements, and lack of interpretability-critical barriers to clinical adoption. We present EAGLE (Efficient Alignment of Generalized Latent Embeddings), a novel deep learning framework that addresses these limitati…

Decoding Dense Embeddings: Sparse Autoencoders for Interpreting and Discretizing Dense Retrieval
Seongwan Park, Taeklim Kim, Youngjoong Ko
https://arxiv.org/abs/2506.00041

Decoding Dense Embeddings: Sparse Autoencoders for Interpreting and Discretizing Dense Retrieval
Despite their strong performance, Dense Passage Retrieval (DPR) models suffer from a lack of interpretability. In this work, we propose a novel interpretability framework that leverages Sparse Autoencoders (SAEs) to decompose previously uninterpretable dense embeddings from DPR models into distinct, interpretable latent concepts. We generate natural language descriptions for each latent concept, enabling human interpretations of both the dense embeddings and the query-document similarity scores…

Spectral coverings without embeddings
Eric Boulter, Steven Rayan
https://arxiv.org/abs/2507.02127 https://arxiv.org/pdf/2507.02127

Spectral coverings without embeddings
In this article, we investigate a weakened version of the spectral correspondence for twisted Higgs bundles. Namely, we construct twisted Higgs bundles from a finite covering map and a vector bundle on that covering but without requiring that they match the eigen-data for some fixed twisted Higgs bundle. We investigate stability for twisted Higgs bundles constructed in this way, and compare our covering data to that of the traditional spectral cover.

Evaluating the Effectiveness of Pre-Trained Audio Embeddings for Classification of Parkinson's Disease Speech Data
Emmy Postma, Cristian Tejedor-Garcia
https://arxiv.org/abs/2506.02078

Evaluating the Effectiveness of Pre-Trained Audio Embeddings for Classification of Parkinson's Disease Speech Data
Speech impairments are prevalent biomarkers for Parkinson's Disease (PD), motivating the development of diagnostic techniques using speech data for clinical applications. Although deep acoustic features have shown promise for PD classification, their effectiveness often varies due to individual speaker differences, a factor that has not been thoroughly explored in the existing literature. This study investigates the effectiveness of three pre-trained audio embeddings (OpenL3, VGGish and Wav2Vec…

Hierarchical Interaction Summarization and Contrastive Prompting for Explainable Recommendations
Yibin Liu, Ang Li, Shijian Li
https://arxiv.org/abs/2507.06044

Hierarchical Interaction Summarization and Contrastive Prompting for Explainable Recommendations
Explainable recommendations, which use the information of user and item with interaction to generate a explanation for why the user would interact with the item, are crucial for improving user trust and decision transparency to the recommender system. Existing methods primarily rely on encoding features of users and items to embeddings, which often leads to information loss due to dimensionality reduction, sparse interactions, and so on. With the advancements of large language models (LLMs) in …

This https://arxiv.org/abs/2410.16428 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSD_…

Neural Scoring: A Refreshed End-to-End Approach for Speaker Recognition in Complex Conditions
Modern speaker verification systems primarily rely on speaker embeddings and cosine similarity. While effective, these methods struggle with multi-talker speech due to the unidentifiability of embedding vectors. We propose Neural Scoring (NS), a novel end-to-end framework that directly estimates verification posterior probabilities without relying on test-side embeddings, making it more powerful and robust to complex conditions, e.g., with multiple talkers. To address the challenge of training …

Semantic Certainty Assessment in Vector Retrieval Systems: A Novel Framework for Embedding Quality Evaluation
Y. Du
https://arxiv.org/abs/2507.05933 https:…

Semantic Certainty Assessment in Vector Retrieval Systems: A Novel Framework for Embedding Quality Evaluation
Vector retrieval systems exhibit significant performance variance across queries due to heterogeneous embedding quality. We propose a lightweight framework for predicting retrieval performance at the query level by combining quantization robustness and neighborhood density metrics. Our approach is motivated by the observation that high-quality embeddings occupy geometrically stable regions in the embedding space and exhibit consistent neighborhood structures. We evaluate our method on 4 standar…

On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Survey
Meishan Zhang, Xin Zhang, Xinping Zhao, Shouzheng Huang, Baotian Hu, Min Zhang
https://arxiv.org/abs/2507.20783

On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Survey
Text embeddings have attracted growing interest due to their effectiveness across a wide range of natural language processing (NLP) tasks, such as retrieval, classification, clustering, bitext mining, and summarization. With the emergence of pretrained language models (PLMs), general-purpose text embeddings (GPTE) have gained significant traction for their ability to produce rich, transferable representations. The general architecture of GPTE typically leverages PLMs to derive dense text repres…

Universal embeddings of flag manifolds and rigidity phenomena
Andrea Loi, Roberto Mossa, Fabio Zuddas
https://arxiv.org/abs/2507.23606 https://arxiv.org/pd…

Universal embeddings of flag manifolds and rigidity phenomena
We prove a universal embedding theorem for flag manifolds: every flag manifold admits a holomorphic isometric embedding into an irreducible classical flag manifold. This result generalizes the classical celebrated embedding theorems of Takeuchi [30] and Nakagawa-Takagi [27]. Using this embedding, we establish new rigidity phenomena for holomorphic isometries between homogeneous Kähler manifolds. As a first immediate consequence we show the triviality of a Kähler-Ricci soliton submanifod of $C…

Do Recommender Systems Really Leverage Multimodal Content? A Comprehensive Analysis on Multimodal Representations for Recommendation
Claudio Pomo, Matteo Attimonelli, Danilo Danese, Fedelucio Narducci, Tommaso Di Noia
https://arxiv.org/abs/2508.04571

Do Recommender Systems Really Leverage Multimodal Content? A Comprehensive Analysis on Multimodal Representations for Recommendation
Multimodal Recommender Systems aim to improve recommendation accuracy by integrating heterogeneous content, such as images and textual metadata. While effective, it remains unclear whether their gains stem from true multimodal understanding or increased model complexity. This work investigates the role of multimodal item embeddings, emphasizing the semantic informativeness of the representations. Initial experiments reveal that embeddings from standard extractors (e.g., ResNet50, Sentence-Bert)…

Verified Language Processing with Hybrid Explainability: A Technical Report
Oliver Robert Fox, Giacomo Bergami, Graham Morgan
https://arxiv.org/abs/2507.05017

Verified Language Processing with Hybrid Explainability: A Technical Report
The volume and diversity of digital information have led to a growing reliance on Machine Learning techniques, such as Natural Language Processing, for interpreting and accessing appropriate data. While vector and graph embeddings represent data for similarity tasks, current state-of-the-art pipelines lack guaranteed explainability, failing to determine similarity for given full texts accurately. These considerations can also be applied to classifiers exploiting generative language models with …

LAPS-Diff: A Diffusion-Based Framework for Singing Voice Synthesis With Language Aware Prosody-Style Guided Learning
Sandipan Dhar, Mayank Gupta, Preeti Rao
https://arxiv.org/abs/2507.04966

LAPS-Diff: A Diffusion-Based Framework for Singing Voice Synthesis With Language Aware Prosody-Style Guided Learning
The field of Singing Voice Synthesis (SVS) has seen significant advancements in recent years due to the rapid progress of diffusion-based approaches. However, capturing vocal style, genre-specific pitch inflections, and language-dependent characteristics remains challenging, particularly in low-resource scenarios. To address this, we propose LAPS-Diff, a diffusion model integrated with language-aware embeddings and a vocal-style guided learning mechanism, specifically designed for Bollywood Hin…

A Dynamic Framework for Semantic Grouping of Common Data Elements (CDE) Using Embeddings and Clustering
Madan Krishnamurthy, Daniel Korn, Melissa A Haendel, Christopher J Mungall, Anne E Thessen
https://arxiv.org/abs/2506.02160

A Dynamic Framework for Semantic Grouping of Common Data Elements (CDE) Using Embeddings and Clustering
This research aims to develop a dynamic and scalable framework to facilitate harmonization of Common Data Elements (CDEs) across heterogeneous biomedical datasets by addressing challenges such as semantic heterogeneity, structural variability, and context dependence to streamline integration, enhance interoperability, and accelerate scientific discovery. Our methodology leverages Large Language Models (LLMs) for context-aware text embeddings that convert CDEs into dense vectors capturing semant…

Soft Injection of Task Embeddings Outperforms Prompt-Based In-Context Learning
Jungwon Park, Wonjong Rhee
https://arxiv.org/abs/2507.20906 https://arxiv.or…

Soft Injection of Task Embeddings Outperforms Prompt-Based In-Context Learning
In-Context Learning (ICL) enables Large Language Models (LLMs) to perform tasks by conditioning on input-output examples in the prompt, without requiring any update in model parameters. While widely adopted, it remains unclear whether prompting with multiple examples is the most effective and efficient way to convey task information. In this work, we propose Soft Injection of task embeddings. The task embeddings are constructed only once using few-shot ICL prompts and repeatedly used during inf…

This https://arxiv.org/abs/2504.06212 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

NNN: Next-Generation Neural Networks for Marketing Measurement
We present NNN, an experimental Transformer-based neural network approach to marketing measurement. Unlike Marketing Mix Models (MMMs) which rely on scalar inputs and parametric decay functions, NNN uses rich embeddings to capture both quantitative and qualitative aspects of marketing and organic channels (e.g., search queries, ad creatives). This, combined with its attention mechanism, potentially enables NNN to model complex interactions, capture long-term effects, and improve sales attributi…

Word stress in self-supervised speech models: A cross-linguistic comparison
Martijn Bentum, Louis ten Bosch, Tomas O. Lentz
https://arxiv.org/abs/2507.04738

Word stress in self-supervised speech models: A cross-linguistic comparison
In this paper we study word stress representations learned by self-supervised speech models (S3M), specifically the Wav2vec 2.0 model. We investigate the S3M representations of word stress for five different languages: Three languages with variable or lexical stress (Dutch, English and German) and two languages with fixed or demarcative stress (Hungarian and Polish). We train diagnostic stress classifiers on S3M embeddings and show that they can distinguish between stressed and unstressed sylla…

Analyzing and Improving Speaker Similarity Assessment for Speech Synthesis
Marc-Andr\'e Carbonneau, Benjamin van Niekerk, Hugo Seut\'e, Jean-Philippe Letendre, Herman Kamper, Julian Za\"idi
https://arxiv.org/abs/2507.02176

Analyzing and Improving Speaker Similarity Assessment for Speech Synthesis
Modeling voice identity is challenging due to its multifaceted nature. In generative speech systems, identity is often assessed using automatic speaker verification (ASV) embeddings, designed for discrimination rather than characterizing identity. This paper investigates which aspects of a voice are captured in such representations. We find that widely used ASV embeddings focus mainly on static features like timbre and pitch range, while neglecting dynamic elements such as rhythm. We also ident…

Rethinking Hybrid Retrieval: When Small Embeddings and LLM Re-ranking Beat Bigger Models
Arjun Rao, Hanieh Alipour, Nick Pendar
https://arxiv.org/abs/2506.00049

Rethinking Hybrid Retrieval: When Small Embeddings and LLM Re-ranking Beat Bigger Models
This paper presents a comparison of embedding models in tri-modal hybrid retrieval for Retrieval-Augmented Generation (RAG) systems. We investigate the fusion of dense semantic, sparse lexical, and graph-based embeddings, focusing on the performance of the MiniLM-v6 and BGE-Large architectures. Contrary to conventional assumptions, our results show that the compact MiniLM-v6 outperforms the larger BGE-Large when integrated with LLM-based re-ranking within our tri-modal hybrid framework. Experim…

Resource-Efficient Adaptation of Large Language Models for Text Embeddings via Prompt Engineering and Contrastive Fine-tuning
Benedikt Roth, Stephan Rappensperger, Tianming Qiu, Hamza Imamovi\'c, Julian W\"ormann, Hao Shen
https://arxiv.org/abs/2507.22729

Resource-Efficient Adaptation of Large Language Models for Text Embeddings via Prompt Engineering and Contrastive Fine-tuning
Large Language Models (LLMs) have become a cornerstone in Natural Language Processing (NLP), achieving impressive performance in text generation. Their token-level representations capture rich, human-aligned semantics. However, pooling these vectors into a text embedding discards crucial information. Nevertheless, many non-generative downstream tasks, such as clustering, classification, or retrieval, still depend on accurate and controllable sentence- or document-level embeddings. We explore se…

Improving Audio Classification by Transitioning from Zero- to Few-Shot
James Taylor, Wolfgang Mack
https://arxiv.org/abs/2507.20036 https://arxiv.org/pdf/2…

Improving Audio Classification by Transitioning from Zero- to Few-Shot
State-of-the-art audio classification often employs a zero-shot approach, which involves comparing audio embeddings with embeddings from text describing the respective audio class. These embeddings are usually generated by neural networks trained through contrastive learning to align audio and text representations. Identifying the optimal text description for an audio class is challenging, particularly when the class comprises a wide variety of sounds. This paper examines few-shot methods desig…

Interact2Vec -- An efficient neural network-based model for simultaneously learning users and items embeddings in recommender systems
Pedro R. Pires, Tiago A. Almeida
https://arxiv.org/abs/2506.22648

Interact2Vec -- An efficient neural network-based model for simultaneously learning users and items embeddings in recommender systems
Over the past decade, recommender systems have experienced a surge in popularity. Despite notable progress, they grapple with challenging issues, such as high data dimensionality and sparseness. Representing users and items as low-dimensional embeddings learned via neural networks has become a leading solution. However, while recent studies show promising results, many approaches rely on complex architectures or require content data, which may not always be available. This paper presents Intera…

Evolutionary Feature-wise Thresholding for Binary Representation of NLP Embeddings
Soumen Sinha, Shahryar Rahnamayan, Azam Asilian Bidgoli
https://arxiv.org/abs/2507.17025

Evolutionary Feature-wise Thresholding for Binary Representation of NLP Embeddings
Efficient text embedding is crucial for large-scale natural language processing (NLP) applications, where storage and computational efficiency are key concerns. In this paper, we explore how using binary representations (barcodes) instead of real-valued features can be used for NLP embeddings derived from machine learning models such as BERT. Thresholding is a common method for converting continuous embeddings into binary representations, often using a fixed threshold across all features. We pr…

Semantic IDs for Music Recommendation
M. Jeffrey Mei, Florian Henkel, Samuel E. Sandberg, Oliver Bembom, Andreas F. Ehmann
https://arxiv.org/abs/2507.18800 https://

Semantic IDs for Music Recommendation
Training recommender systems for next-item recommendation often requires unique embeddings to be learned for each item, which may take up most of the trainable parameters for a model. Shared embeddings, such as using content information, can reduce the number of distinct embeddings to be stored in memory. This allows for a more lightweight model; correspondingly, model complexity can be increased due to having fewer embeddings to store in memory. We show the benefit of using shared content-base…

Intrinsic vs. Extrinsic Evaluation of Czech Sentence Embeddings: Semantic Relevance Doesn't Help with MT Evaluation
Petra Baran\v{c}\'ikov\'a, Ond\v{r}ej Bojar
https://arxiv.org/abs/2506.20203

Intrinsic vs. Extrinsic Evaluation of Czech Sentence Embeddings: Semantic Relevance Doesn't Help with MT Evaluation
In this paper, we compare Czech-specific and multilingual sentence embedding models through intrinsic and extrinsic evaluation paradigms. For intrinsic evaluation, we employ Costra, a complex sentence transformation dataset, and several Semantic Textual Similarity (STS) benchmarks to assess the ability of the embeddings to capture linguistic phenomena such as semantic similarity, temporal aspects, and stylistic variations. In the extrinsic evaluation, we fine-tune each embedding model using COM…

Context is Gold to find the Gold Passage: Evaluating and Training Contextual Document Embeddings
Max Conti, Manuel Faysse, Gautier Viaud, Antoine Bosselut, C\'eline Hudelot, Pierre Colombo
https://arxiv.org/abs/2505.24782

Context is Gold to find the Gold Passage: Evaluating and Training Contextual Document Embeddings
A limitation of modern document retrieval embedding methods is that they typically encode passages (chunks) from the same documents independently, often overlooking crucial contextual information from the rest of the document that could greatly improve individual chunk representations. In this work, we introduce ConTEB (Context-aware Text Embedding Benchmark), a benchmark designed to evaluate retrieval models on their ability to leverage document-wide context. Our results show that state-of-t…

ViLLA-MMBench: A Unified Benchmark Suite for LLM-Augmented Multimodal Movie Recommendation
Fatemeh Nazary, Ali Tourani, Yashar Deldjoo, Tommaso Di Noia
https://arxiv.org/abs/2508.04206

ViLLA-MMBench: A Unified Benchmark Suite for LLM-Augmented Multimodal Movie Recommendation
Recommending long-form video content demands joint modeling of visual, audio, and textual modalities, yet most benchmarks address only raw features or narrow fusion. We present ViLLA-MMBench, a reproducible, extensible benchmark for LLM-augmented multimodal movie recommendation. Built on MovieLens and MMTF-14K, it aligns dense item embeddings from three modalities: audio (block-level, i-vector), visual (CNN, AVF), and text. Missing or sparse metadata is automatically enriched using state-of-the…

SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts
Marc Brinner, Sina Zarriess
https://arxiv.org/abs/2507.13105

SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts
We introduce SemCSE, an unsupervised method for learning semantic embeddings of scientific texts. Building on recent advances in contrastive learning for text embeddings, our approach leverages LLM-generated summaries of scientific abstracts to train a model that positions semantically related summaries closer together in the embedding space. This resulting objective ensures that the model captures the true semantic content of a text, in contrast to traditional citation-based approaches that do…

Unsupervised Document and Template Clustering using Multimodal Embeddings
Phillipe R. Sampaio, Helene Maxcici
https://arxiv.org/abs/2506.12116 https://

Unsupervised Document and Template Clustering using Multimodal Embeddings
This paper investigates a novel approach to unsupervised document clustering by leveraging multimodal embeddings as input to traditional clustering algorithms such as $k$-Means and DBSCAN. Our method aims to achieve a finer-grained document understanding by not only grouping documents at the type level (e.g., invoices, purchase orders), but also distinguishing between different templates within the same document category. This is achieved by using embeddings that capture textual content, layout…