Tootfinder

ILT-Iterative LoRA Training through Focus-Feedback-Fix for Multilingual Speech Recognition
Qingliang Meng, Hao Wu, Wei Liang, Wei Xu, Qing Zhao
https://arxiv.org/abs/2507.08477

ILT-Iterative LoRA Training through Focus-Feedback-Fix for Multilingual Speech Recognition
The deep integration of large language models and automatic speech recognition systems has become a promising research direction with high practical value. To address the overfitting issue commonly observed in Low-Rank Adaptation (LoRA) during the supervised fine-tuning (SFT) stage, this work proposes an innovative training paradigm Iterative LoRA Training (ILT) in combination with an Iterative Pseudo Labeling strategy, effectively enhancing the theoretical upper bound of model performance. Bas…

Inverse Synthetic Aperture Radar, Radar Cross Section, and Iterative Smooth Reweighting $\ell_1$-minimization
Christer Larsson
https://arxiv.org/abs/2508.10536 https://

Inverse Synthetic Aperture Radar, Radar Cross Section, and Iterative Smooth Reweighting $\ell_1$-minimization
Radar Cross Section measurement data is often analyzed using Inverse Synthetic Aperture Radar images. This paper compares backprojection and iterative smooth reweighted $\ell_1$-minimization as methods to analyze radar cross section measurements and extract radar cross section for parts of the measured object. The main conclusion is that using backprojection images to extract RCS is robust and accurate but is more limited by the resolution than iterative smooth reweighted $\ell_1$-minimization.…

On the Consistency and Performance of the Iterative Bayesian Update
Ehab ElSalamouny, Catuscia Palamidessi
https://arxiv.org/abs/2508.09980 https://arxiv.o…

On the Consistency and Performance of the Iterative Bayesian Update
For many social, scientific, and commercial purposes, it is often important to estimate the distribution of the users' data regarding a sensitive attribute, e.g., their ages, locations, etc. To allow this estimation while protecting the users' privacy, every user applies a local privacy protection mechanism that releases a noisy (sanitized) version of their original datum to the data collector; then the original distribution is estimated using one of the known methods, such as the matrix invers…

An Iterative Algorithm for Differentially Private $k$-PCA with Adaptive Noise
Johanna D\"ungler, Amartya Sanyal
https://arxiv.org/abs/2508.10879 https://

An Iterative Algorithm for Differentially Private $k$-PCA with Adaptive Noise
Given $n$ i.i.d. random matrices $A_i \in \mathbb{R}^{d \times d}$ that share a common expectation $Σ$, the objective of Differentially Private Stochastic PCA is to identify a subspace of dimension $k$ that captures the largest variance directions of $Σ$, while preserving differential privacy (DP) of each individual $A_i$. Existing methods either (i) require the sample size $n$ to scale super-linearly with dimension $d$, even under Gaussian assumptions on the $A_i$, or (ii) introduce excessiv…

The Minnesota assassinations, the attempted arson at Governor Shapiro's home, and the violent arrest of Senator Padilla are all straight out of the Jim Crow terrorism playbook. Same goes with cops aiming for journalists covering the ICE protests.
None of these are new tactics, just a new iteration.
-- Max Kennerly

Verify Distributed Deep Learning Model Implementation Refinement with Iterative Relation Inference
Zhanghan Wang, Ding Ding, Hang Zhu, Haibin Lin, Aurojit Panda
https://arxiv.org/abs/2508.09505

Verify Distributed Deep Learning Model Implementation Refinement with Iterative Relation Inference
Distributed machine learning training and inference is common today because today's large models require more memory and compute than can be provided by a single GPU. Distributed models are generally produced by programmers who take a sequential model specification and apply several distribution strategies to distribute state and computation across GPUs. Unfortunately, bugs can be introduced in the process, and a distributed model implementation's outputs might differ from the sequential model'…

CUE-RAG: Towards Accurate and Cost-Efficient Graph-Based RAG via Multi-Partite Graph and Query-Driven Iterative Retrieval
Yaodong Su, Yixiang Fang, Yingli Zhou, Quanqing Xu, Chuanhui Yang
https://arxiv.org/abs/2507.08445

CUE-RAG: Towards Accurate and Cost-Efficient Graph-Based RAG via Multi-Partite Graph and Query-Driven Iterative Retrieval
Despite the remarkable progress of Large Language Models (LLMs), their performance in question answering (QA) remains limited by the lack of domain-specific and up-to-date knowledge. Retrieval-Augmented Generation (RAG) addresses this limitation by incorporating external information, often from graph-structured data. However, existing graph-based RAG methods suffer from poor graph quality due to incomplete extraction and insufficient utilization of query information during retrieval. To overcom…

Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials
Ekaterina Grishina, Matvey Smirnov, Maxim Rakhuba
https://arxiv.org/abs/2506.10935

Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials
The problem of computing optimal orthogonal approximation to a given matrix has attracted growing interest in machine learning. Notable applications include the recent Muon optimizer or Riemannian optimization on the Stiefel manifold. Among existing approaches, the Newton-Schulz iteration has emerged as a particularly effective solution, as it relies solely on matrix multiplications and thus achieves high computational efficiency on GPU hardware. Despite its efficiency, the method has inherent …

Dimension-Free Bounds for Generalized First-Order Methods via Gaussian Coupling
Galen Reeves
https://arxiv.org/abs/2508.10782 https://arxiv.org/pdf/2508.10…

Dimension-Free Bounds for Generalized First-Order Methods via Gaussian Coupling
We establish non-asymptotic bounds on the finite-sample behavior of generalized first-order iterative algorithms -- including gradient-based optimization methods and approximate message passing (AMP) -- with Gaussian data matrices and full-memory, non-separable nonlinearities. The central result constructs an explicit coupling between the iterates of a generalized first-order method and a conditionally Gaussian process whose covariance evolves deterministically via a finite-dimensional state ev…

Information Retention in Iterative Random Projection of Convex Bodies to Lower Dimensions
Nazanin Mirhosseini
https://arxiv.org/abs/2508.10218 https://arxi…

Information Retention in Iterative Random Projection of Convex Bodies to Lower Dimensions
In this paper, we consider a bounded convex body $K_0 \subset \mathbb{R}^{n}$ subjected to two successive random orthogonal projections onto $\mathbb{R}^{n-1}$ and $\mathbb{R}^{n-2}$, respectively. First, we project $K_0$ orthogonally onto $U_{1}^{\perp}$, the orthogonal complement of $\mbox{\boldmath $U$}_1$, where $\mbox{\boldmath $U$}_1$ is uniformly distributed on the unit sphere $S^{n-1}$. This yields a random convex body $K_1 = \mathrm{Proj}_{{U_1}^{\perp}}(K_0) \subset \mathbb{R}^{n-1}$.…