Search:
Match:
16 results
Research Paper#Time Series Analysis, Matrix Factorization, Machine Learning🔬 ResearchAnalyzed: Jan 3, 2026 06:13

Modewise Additive Factor Model for Matrix Time Series

Published:Dec 31, 2025 18:24
1 min read
ArXiv

Analysis

This paper introduces a novel Modewise Additive Factor Model (MAFM) for matrix-valued time series, offering a more flexible approach than existing multiplicative factor models like Tucker and CP. The key innovation lies in its additive structure, allowing for separate modeling of row-specific and column-specific latent effects. The paper's contribution is significant because it provides a computationally efficient estimation procedure (MINE and COMPAS) and a data-driven inference framework, including convergence rates, asymptotic distributions, and consistent covariance estimators. The development of matrix Bernstein inequalities for quadratic forms of dependent matrix time series is a valuable technical contribution. The paper's focus on matrix time series analysis is relevant to various fields, including finance, signal processing, and recommendation systems.
Reference

“The key methodological innovation is that orthogonal complement projections completely eliminate cross-modal interference when estimating each loading space.”

PermalinkArXiv
Research Paper#Machine Learning, Matrix Factorization, Count Data, Overdispersion🔬 ResearchAnalyzed: Jan 3, 2026 08:53

Generalized Poisson NMF for Overdispersed Count Data

Published:Dec 31, 2025 03:51
1 min read
ArXiv

Analysis

This paper addresses the limitations of existing Non-negative Matrix Factorization (NMF) models, specifically those based on Poisson and Negative Binomial distributions, when dealing with overdispersed count data. The authors propose a new NMF model using the Generalized Poisson distribution, which offers greater flexibility in handling overdispersion and improves the applicability of NMF to a wider range of count data scenarios. The core contribution is the introduction of a maximum likelihood approach for parameter estimation within this new framework.
Reference

“The paper proposes a non-negative matrix factorization based on the generalized Poisson distribution, which can flexibly accommodate overdispersion, and introduces a maximum likelihood approach for parameter estimation.”

PermalinkArXiv
Paper#llm🔬 ResearchAnalyzed: Jan 3, 2026 09:22

Multi-Envelope DBF for LLM Quantization

Published:Dec 31, 2025 01:04
1 min read
ArXiv

Analysis

This paper addresses the limitations of Double Binary Factorization (DBF) for extreme low-bit quantization of Large Language Models (LLMs). DBF, while efficient, suffers from performance saturation due to restrictive scaling parameters. The proposed Multi-envelope DBF (MDBF) improves upon DBF by introducing a rank-$l$ envelope, allowing for better magnitude expressiveness while maintaining a binary carrier and deployment-friendly inference. The paper demonstrates improved perplexity and accuracy on LLaMA and Qwen models.
Reference

“MDBF enhances perplexity and zero-shot accuracy over previous binary formats at matched bits per weight while preserving the same deployment-friendly inference primitive.”

PermalinkArXiv
Research Paper#Functional Analysis, Operator Theory🔬 ResearchAnalyzed: Jan 3, 2026 15:36

Functional Models for Gamma-n Contractions

Published:Dec 30, 2025 17:03
1 min read
ArXiv

Analysis

This paper explores functional models for Γ_n-contractions, building upon existing models for contractions. It aims to provide a deeper understanding of these operators through factorization and model construction, potentially leading to new insights into their behavior and properties. The paper's significance lies in extending the theory of contractions to a more general class of operators.
Reference

“The paper establishes factorization results that clarify the relationship between a minimal isometric dilation and an arbitrary isometric dilation of a contraction.”

PermalinkArXiv
Research Paper#Deep Learning, Backpropagation, KL Divergence, Probabilistic Inference🔬 ResearchAnalyzed: Jan 3, 2026 17:14

Backpropagation and KL Projections: Exact Correspondences

Published:Dec 30, 2025 16:42
1 min read
ArXiv

Analysis

This paper explores the mathematical connections between backpropagation, a core algorithm in deep learning, and Kullback-Leibler (KL) divergence, a measure of the difference between probability distributions. It establishes two precise relationships, showing that backpropagation can be understood through the lens of KL projections. This provides a new perspective on how backpropagation works and potentially opens avenues for new algorithms or theoretical understanding. The focus on exact correspondences is significant, as it provides a strong mathematical foundation.
Reference

“Backpropagation arises as the differential of a KL projection map on a delta-lifted factorization.”

PermalinkArXiv
Research Paper#Computer Vision, Image Processing, Intrinsic Image Decomposition, Transformers🔬 ResearchAnalyzed: Jan 3, 2026 16:01

IDT: Multi-View Intrinsic Decomposition with a Physically Grounded Transformer

Published:Dec 29, 2025 18:24
1 min read
ArXiv

Analysis

This paper introduces IDT, a novel feed-forward transformer-based framework for multi-view intrinsic image decomposition. It addresses the challenge of view inconsistency in existing methods by jointly reasoning over multiple input images. The use of a physically grounded image formation model, decomposing images into diffuse reflectance, diffuse shading, and specular shading, is a key contribution, enabling interpretable and controllable decomposition. The focus on multi-view consistency and the structured factorization of light transport are significant advancements in the field.
Reference

“IDT produces view-consistent intrinsic factors in a single forward pass, without iterative generative sampling.”

PermalinkArXiv
Research Paper#Astronomy/Exoplanets🔬 ResearchAnalyzed: Jan 3, 2026 16:16

Shot-Noise-Limited Radial Velocity Extraction via Spectral Factorization

Published:Dec 28, 2025 18:56
1 min read
ArXiv

Analysis

This paper presents a novel method for extracting radial velocities from spectroscopic data, achieving high precision by factorizing the data into principal spectra and time-dependent kernels. This approach allows for the recovery of both spectral components and radial velocity shifts simultaneously, leading to improved accuracy, especially in the presence of spectral variability. The validation on synthetic and real-world datasets, including observations of HD 34411 and τ Ceti, demonstrates the method's effectiveness and its ability to reach the instrumental precision limit. The ability to detect signals with semi-amplitudes down to ~50 cm/s is a significant advancement in the field of exoplanet detection.
Reference

“The method recovers coherent signals and reaches the instrumental precision limit of ~30 cm/s.”

PermalinkArXiv
research#mathematics🔬 ResearchAnalyzed: Jan 4, 2026 06:50

Pita factorisation in operadic categories

Published:Dec 28, 2025 05:36
1 min read
ArXiv

Analysis

This article likely discusses a mathematical concept related to category theory and operads. The title suggests a specific factorization technique ('Pita factorisation') within the context of operadic categories. The source, ArXiv, indicates this is a pre-print or research paper.

Key Takeaways

    Reference

    “”

    PermalinkArXiv
    Research Paper#Nuclear Physics, Neutrino Physics, Particle Physics🔬 ResearchAnalyzed: Jan 3, 2026 19:37

    Factorized Calculations for Particle Interactions in Nuclei

    Published:Dec 28, 2025 03:40
    1 min read
    ArXiv

    Analysis

    This paper proposes a factorized approach to calculate nuclear currents, simplifying calculations for electron, neutrino, and beyond Standard Model (BSM) processes. The factorization separates nucleon dynamics from nuclear wave function overlaps, enabling efficient computation and flexible modification of nucleon couplings. This is particularly relevant for event generators used in neutrino physics and other areas where accurate modeling of nuclear effects is crucial.
    Reference

    “The factorized form is attractive for (neutrino) event generators: it abstracts away the nuclear model and allows to easily modify couplings to the nucleon.”

    PermalinkArXiv
    Research Paper#Theoretical Physics, Celestial Holography, Gauge Theory🔬 ResearchAnalyzed: Jan 3, 2026 20:10

    Celestial Theory Constraints from Soft Radiation Data

    Published:Dec 26, 2025 18:17
    1 min read
    ArXiv

    Analysis

    This paper analyzes high-order gauge-theory calculations, translated into celestial language, to test and constrain celestial holography. It focuses on soft emission currents and their implications for the celestial theory, particularly questioning the need for a logarithmic celestial theory and exploring the structure of multiple emission currents.
    Reference

    “All logarithms arising in the loop expansion of the single soft current can be reabsorbed in the scale choices for the $d$-dimensional coupling, casting some doubt on the need for a logarithmic celestial theory.”

    PermalinkArXiv
    Research#llm🔬 ResearchAnalyzed: Jan 4, 2026 08:40

    Weighted Fourier Factorizations: Optimal Gaussian Noise for Differentially Private Marginal and Product Queries

    Published:Dec 25, 2025 04:14
    1 min read
    ArXiv

    Analysis

    This article, sourced from ArXiv, likely presents a novel approach to differentially private data analysis. The title suggests a focus on optimizing the addition of Gaussian noise, a common technique for achieving differential privacy, in the context of marginal and product queries. The use of "Weighted Fourier Factorizations" indicates a potentially sophisticated mathematical framework. The research likely aims to improve the accuracy and utility of private data analysis by minimizing the noise added while still maintaining privacy guarantees.
    Reference

    “”

    PermalinkArXiv
    Research#llm🔬 ResearchAnalyzed: Jan 4, 2026 08:36

    On-shell representation and further instances of the 2-split behavior of amplitudes

    Published:Dec 23, 2025 21:37
    1 min read
    ArXiv

    Analysis

    This article likely discusses advanced topics in theoretical physics, specifically focusing on the behavior of amplitudes in particle physics. The title suggests an exploration of how these amplitudes can be represented and how they exhibit a '2-split' behavior, which could relate to factorization properties or other decomposition techniques. The source, ArXiv, indicates this is a peer-reviewed research paper.

    Key Takeaways

      Reference

      “”

      PermalinkArXiv
      Research#llm🔬 ResearchAnalyzed: Jan 4, 2026 10:43

      VALLR-Pin: Uncertainty-Factorized Visual Speech Recognition for Mandarin with Pinyin Guidance

      Published:Dec 23, 2025 03:52
      1 min read
      ArXiv

      Analysis

      This article introduces VALLR-Pin, a new approach to visual speech recognition for Mandarin. The core innovation appears to be the use of uncertainty factorization and Pinyin guidance. The paper likely explores how these techniques improve the accuracy and robustness of the system. The source being ArXiv suggests this is a research paper, focusing on technical details and experimental results.
      Reference

      “”

      PermalinkArXiv
      Research#llm🔬 ResearchAnalyzed: Jan 4, 2026 08:26

      Applying non-negative matrix factorization with covariates to structural equation modeling for blind input-output analysis

      Published:Dec 20, 2025 07:22
      1 min read
      ArXiv

      Analysis

      This article likely presents a novel methodological approach. It combines non-negative matrix factorization (NMF) with structural equation modeling (SEM) and incorporates covariates. The focus is on blind input-output analysis, suggesting applications in areas where the underlying processes are not fully observable. The use of ArXiv indicates it's a pre-print, meaning it's not yet peer-reviewed.
      Reference

      “The article's abstract or introduction would contain the most relevant quote, but without access to the full text, a specific quote cannot be provided.”

      PermalinkArXiv
      Research#Agent🔬 ResearchAnalyzed: Jan 10, 2026 11:51

      Bayesian Factorization for Vision-Language-Action Policies

      Published:Dec 12, 2025 01:59
      1 min read
      ArXiv

      Analysis

      This research paper proposes a novel approach to integrating vision, language, and action within an AI system. The Bayesian factorization method offers a potentially promising way to improve the performance of agents in complex environments.
      Reference

      “The paper focuses on vision-language-action policies.”

      PermalinkArXiv
      Research#llm🔬 ResearchAnalyzed: Dec 25, 2025 04:49

      What exactly does word2vec learn?

      Published:Sep 1, 2025 09:00
      1 min read
      Berkeley AI

      Analysis

      This article from Berkeley AI discusses a new paper that provides a quantitative and predictive theory describing the learning process of word2vec. For years, researchers lacked a solid understanding of how word2vec, a precursor to modern language models, actually learns. The paper demonstrates that in realistic scenarios, the learning problem simplifies to unweighted least-squares matrix factorization. Furthermore, the researchers solved the gradient flow dynamics in closed form, revealing that the final learned representations are essentially derived from PCA. This research sheds light on the inner workings of word2vec and provides a theoretical foundation for understanding its learning dynamics, particularly the sequential, rank-incrementing steps observed during training.
      Reference

      “the final learned representations are simply given by PCA.”

      PermalinkBerkeley AI