Tootfinder

Knowledge-Driven Imitation Learning: Enabling Generalization Across Diverse Conditions
Zhuochen Miao, Jun Lv, Hongjie Fang, Yang Jin, Cewu Lu
https://arxiv.org/abs/2506.21057

Knowledge-Driven Imitation Learning: Enabling Generalization Across Diverse Conditions
Imitation learning has emerged as a powerful paradigm in robot manipulation, yet its generalization capability remains constrained by object-specific dependencies in limited expert demonstrations. To address this challenge, we propose knowledge-driven imitation learning, a framework that leverages external structural semantic knowledge to abstract object representations within the same category. We introduce a novel semantic keypoint graph as a knowledge template and develop a coarse-to-fine te…

Nature: Cortical circuits for cross-modal generalization (in mice) https://www.nature.com/articles/s41467-025-59342-9 "Optogenetic sensory substitution and systematic silencing of these associative areas revealed that a single area in the dorsal stream is necessary an…

Cortical circuits for cross-modal generalization - Nature Communications
Neural mechanisms underlying cross-modal generalization of learned sensorimotor associations are not fully understood. Here authors show that mice use a specialized cortical circuit to generalize learned behaviors between vision and touch. A single region in the dorsal cortex is essential for forming abstract spatial representations that enable cross-modal flexibility.

Stable Minima of ReLU Neural Networks Suffer from the Curse of Dimensionality: The Neural Shattering Phenomenon
Tongtong Liang, Dan Qiao, Yu-Xiang Wang, Rahul Parhi
https://arxiv.org/abs/2506.20779

Stable Minima of ReLU Neural Networks Suffer from the Curse of Dimensionality: The Neural Shattering Phenomenon
We study the implicit bias of flatness / low (loss) curvature and its effects on generalization in two-layer overparameterized ReLU networks with multivariate inputs -- a problem well motivated by the minima stability and edge-of-stability phenomena in gradient-descent training. Existing work either requires interpolation or focuses only on univariate inputs. This paper presents new and somewhat surprising theoretical results for multivariate inputs. On two natural settings (1) generalization g…

Adversarial Training: Enhancing Out-of-Distribution Generalization for Learning Wireless Resource Allocation
Shengjie Liu, Chenyang Yang
https://arxiv.org/abs/2506.21208

Adversarial Training: Enhancing Out-of-Distribution Generalization for Learning Wireless Resource Allocation
Deep neural networks (DNNs) have widespread applications for optimizing resource allocation. Yet, their performance is vulnerable to distribution shifts between training and test data, say channels. In this letter, we resort to adversarial training (AT) for enhancing out-of-distribution (OOD) generalizability of DNNs trained in unsupervised manner. We reformulate AT to capture the OOD degradation, and propose a one-step gradient ascent method for AT. The proposed method is validated by optimizi…

This https://arxiv.org/abs/2502.20991 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLO_…

Domains, Information Frames, Rough Sets: An Equivalence of Categories
A generalization of Scott's information systems~\cite{sco82} is presented that captures exactly all continuous domains. The global consistency predicate in Scott's definition is relativized. Now, for every atomic statement, there is a consistency predicate that states which finite sets of statements express information that is consistent with the given statement. The category of information frames is shown to be equivalent to the category of domains. Moreover, the relationship with CF-approxima…

EA-ViT: Efficient Adaptation for Elastic Vision Transformer
Chen Zhu, Wangbo Zhao, Huiwen Zhang, Samir Khaki, Yuhao Zhou, Weidong Tang, Shuo Wang, Zhihang Yuan, Yuzhang Shang, Xiaojiang Peng, Kai Wang, Dawei Yang
https://arxiv.org/abs/2507.19360

EA-ViT: Efficient Adaptation for Elastic Vision Transformer
Vision Transformers (ViTs) have emerged as a foundational model in computer vision, excelling in generalization and adaptation to downstream tasks. However, deploying ViTs to support diverse resource constraints typically requires retraining multiple, size-specific ViTs, which is both time-consuming and energy-intensive. To address this issue, we propose an efficient ViT adaptation framework that enables a single adaptation process to generate multiple models of varying sizes for deployment on …

Polaritonic Coupled Cluster Theory for Unpolarized Cavities Exploiting Point Group Symmetry
Laurenz Monzel, Stella Stopkowicz
https://arxiv.org/abs/2507.19180 https://

Polaritonic Coupled Cluster Theory for Unpolarized Cavities Exploiting Point Group Symmetry
We introduce a generalization of the quantum electrodynamic coupled cluster (QED-CC)wave function ansatz, to describe the strongly coupled light-matter system in an unpolarized optical Fabry-Pérot cavity. This is achieved by explicitly treating two cavity modes in our calculation with perpendicular polarizations and demonstrate that this ansatz preserves the symmetry of an unpolarized cavity. Furthermore, exploiting point-group symmetry enables the assignment of polaritonic excited states as w…

A generalization of Deodhar's defect statistic for Iwahori--Hecke algebras of type $BC$
Gavin Hobbs, Tommy Parisi, Mark Skandera, Jiayuan Wang
https://arxiv.org/abs/2506.20099

A generalization of Deodhar's defect statistic for Iwahori--Hecke algebras of type $BC$
Let $H$ be the Iwahori--Hecke algebra corresponding to any Coxeter group. Deodhar's defect statistic [Geom. Dedicata 36, (1990) pp.95--119] allows one to expand products of simple Kazhdan--Lusztig basis elements of $H$ in the natural basis of $H$. Clearwater and the third author gave a type-$A$ extension [Ann. Comb. 25, no. 3 (2021) pp.757--787] of this formula which combinatorially describes the natural expansion of products of Kazhdan--Lusztig basis elements indexed by smooth elements of the …

TableMoE: Neuro-Symbolic Routing for Structured Expert Reasoning in Multimodal Table Understanding
Junwen Zhang, Pu Chen, Yin Zhang
https://arxiv.org/abs/2506.21393

TableMoE: Neuro-Symbolic Routing for Structured Expert Reasoning in Multimodal Table Understanding
Multimodal understanding of tables in real-world contexts is challenging due to the complexity of structure, symbolic density, and visual degradation (blur, skew, watermarking, incomplete structures or fonts, multi-span or hierarchically nested layouts). Existing multimodal large language models (MLLMs) struggle with such WildStruct conditions, resulting in limited performance and poor generalization. To address these challenges, we propose TableMoE, a neuro-symbolic Mixture-of-Connector-Expert…

Chern Classes of Toric Variety Bundles
Gregory Taroyan
https://arxiv.org/abs/2506.20848 https://arxiv.org/pdf/2506.20848

Chern Classes of Toric Variety Bundles
We resolve a conjecture of Khovanskii-Monin on the Chern classes of toric variety bundles. The main result is a formula for the total Chern class of the tangent bundle of a toric variety bundle in terms of the total Chern class of the base and the total Chern class of the toric fibre. The result serves as a simultaneous generalization of the description of the total Chern class of a projectivized vector bundle and of the formula for the total Chern class of a toric variety in terms of the Chern…

An isovariant Blakers--Massey theorem
Inbar Klang, Sarah Yeakel
https://arxiv.org/abs/2506.21259 https://arxiv.org/pdf/2506.21259

An isovariant Blakers--Massey theorem
An isovariant map is an equivariant map between $G$-spaces which strictly preserves isotropy groups. In this paper, we lay the groundwork for the study of isovariant stable homotopy theory. We prove an isovariant Blakers--Massey theorem and its $n$-cubical generalization, define a suitable notion of suspension (by a trivial representation sphere) in the isovariant category, and prove an isovariant Freudenthal suspension theorem.

Enhancing Diabetic Retinopathy Classification Accuracy through Dual Attention Mechanism in Deep Learning
Abdul Hannan, Zahid Mahmood, Rizwan Qureshi, Hazrat Ali
https://arxiv.org/abs/2507.19199

Enhancing Diabetic Retinopathy Classification Accuracy through Dual Attention Mechanism in Deep Learning
Automatic classification of Diabetic Retinopathy (DR) can assist ophthalmologists in devising personalized treatment plans, making it a critical component of clinical practice. However, imbalanced data distribution in the dataset becomes a bottleneck in the generalization of deep learning models trained for DR classification. In this work, we combine global attention block (GAB) and category attention block (CAB) into the deep learning model, thus effectively overcoming the imbalanced data dist…

Enhancing Generalization of Spiking Neural Networks Through Temporal Regularization
Boxuan Zhang, Zhen Xu, Kuan Tao
https://arxiv.org/abs/2506.19256 https:…

Enhancing Generalization of Spiking Neural Networks Through Temporal Regularization
Spiking Neural Networks (SNNs) have received widespread attention due to their event-driven and low-power characteristics, making them particularly effective for processing event-based neuromorphic data. Recent studies have shown that directly trained SNNs suffer from severe overfitting issues due to the limited scale of neuromorphic datasets and the gradient mismatching problem, which fundamentally constrain their generalization performance. In this paper, we propose a temporal regularization …

A Multi-Stage Framework for Multimodal Controllable Speech Synthesis
Rui Niu, Weihao Wu, Jie Chen, Long Ma, Zhiyong Wu
https://arxiv.org/abs/2506.20945 htt…

A Multi-Stage Framework for Multimodal Controllable Speech Synthesis
Controllable speech synthesis aims to control the style of generated speech using reference input, which can be of various modalities. Existing face-based methods struggle with robustness and generalization due to data quality constraints, while text prompt methods offer limited diversity and fine-grained control. Although multimodal approaches aim to integrate various modalities, their reliance on fully matched training data significantly constrains their performance and applicability. This pa…

KnowML: Improving Generalization of ML-NIDS with Attack Knowledge Graphs
Xin Fan Guo, Albert Merono Penuela, Sergio Maffeis, Fabio Pierazzi
https://arxiv.org/abs/2506.19802

KnowML: Improving Generalization of ML-NIDS with Attack Knowledge Graphs
Despite extensive research on Machine Learning-based Network Intrusion Detection Systems (ML-NIDS), their capability to detect diverse attack variants remains uncertain. Prior studies have largely relied on homogeneous datasets, which artificially inflate performance scores and offer a false sense of security. Designing systems that can effectively detect a wide range of attack variants remains a significant challenge. The progress of ML-NIDS continues to depend heavily on human expertise, whic…

String Consensus Problems with Swaps and Substitutions
Est\'eban Gabory, Laurent Bulteau, Gabriele Fici, Hilde Verbeek
https://arxiv.org/abs/2507.19139 https://

String Consensus Problems with Swaps and Substitutions
String consensus problems aim at finding a string that minimizes some given distance with respect to an input set of strings. In particular, in the \textsc{Closest String} problem, we are given a set of strings of equal length and a radius $d$. The objective is to find a new string that differs from each input string by at most $d$ substitutions. We study a generalization of this problem where, in addition to substitutions, swaps of adjacent characters are also permitted, each operation incurri…

Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimation
He Li, Haoang Chi, Mingyu Liu, Wanrong Huang, Liyang Xu, Wenjing Yang
https://arxiv.org/abs/2506.21154

Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimation
The real world naturally has dimensions of time and space. Therefore, estimating the counterfactual outcomes with spatial-temporal attributes is a crucial problem. However, previous methods are based on classical statistical models, which still have limitations in performance and generalization. This paper proposes a novel framework for estimating counterfactual outcomes with spatial-temporal attributes using the Transformer, exhibiting stronger estimation ability. Under mild assumptions, the p…

S-r-$ideals in Commutative Rings
Abuzer G\"und\"uz, Osama A. Naji, Mehmet \"Ozen
https://arxiv.org/abs/2505.19198 https://

S-r-$ideals in Commutative Rings
This article introduces the notion of $S-r-$ideals in commutative ring $H$ as a generalization of $ r-$ideals. Let $S$ be a multiplicatively closed subset and $A$ an ideal of $H$ with $S\cap A= \emptyset$. $A$ is called $S-r-ideal$ if $ \exists s\in S$ such that for $w,z\in H,\; wz\in A$ and $Ann(w)=0$, then $sz\in A.$ Basic properties of $S-r-$ideals are given. It is shown that an $r-$ideal is always an $S-r-$ideal, and the converse is true under some conditions. Various characterizations of $…

After training, we finetune on real-world data. We observe that the models that have been pre-trained with noise converge very quickly compared to a baseline which is trained from scratch.
Moreover, on the other datasets, the UP models retain their zero-shot performance during finetuning. This suggests that there may be a generalization benefit to using a UP model.
All this is at the expense of much longer training, but that cost can be amortized over many tasks.

Replaced article(s) found for math.QA. https://arxiv.org/list/math.QA/new
[1/1]:
- A Non-Abelian Generalization of the Alexander Polynomial from Quantum $\mathfrak{sl}_3$
Matthew Harper

Detecting weighted hidden cliques
Urmisha Chatterjee, Karissa Huang, Ritabrata Karmakar, B. R. Vinay Kumar, G\'abor Lugosi, Nandan Malhotra, Anirban Mandal, Maruf Alam Tarafdar
https://arxiv.org/abs/2506.21543

Detecting weighted hidden cliques
We study a generalization of the classical hidden clique problem to graphs with real-valued edge weights. Formally, we define a hypothesis testing problem. Under the null hypothesis, edges of a complete graph on $n$ vertices are associated with independent and identically distributed edge weights from a distribution $P$. Under the alternate hypothesis, $k$ vertices are chosen at random and the edge weights between them are drawn from a distribution $Q$, while the remaining are sampled from $P$.…

This https://arxiv.org/abs/2412.17187 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

Generalized Homogeneous Derivations of Graded Rings
In this manuscript, we introduce a novel concept in graded rings called generalized homogeneous derivations, which serve as a natural generalization of the homogeneous derivations introduced by Kanunnikov. We establish the specialized notion of gr-generalized derivations, a subclass that preserves the degree of homogeneous elements. We extend several significant results, originally established for prime rings, to the context of gr-prime rings. Furthermore, we characterize when gr-semiprime ring…

A DPI-PAC-Bayesian Framework for Generalization Bounds
Muhan Guan, Farhad Farokhi, Jingge Zhu
https://arxiv.org/abs/2507.14795 https://

A DPI-PAC-Bayesian Framework for Generalization Bounds
We develop a unified Data Processing Inequality PAC-Bayesian framework -- abbreviated DPI-PAC-Bayesian -- for deriving generalization error bounds in the supervised learning setting. By embedding the Data Processing Inequality (DPI) into the change-of-measure technique, we obtain explicit bounds on the binary Kullback-Leibler generalization gap for both Rényi divergence and any $f$-divergence measured between a data-independent prior distribution and an algorithm-dependent posterior distributi…

Precise Gradient Discontinuities in Neural Fields for Subspace Physics
Mengfei Liu, Yue Chang, Zhecheng Wang, Peter Yichen Chen, Eitan Grinspun
https://arxiv.org/abs/2505.20421

Precise Gradient Discontinuities in Neural Fields for Subspace Physics
Discontinuities in spatial derivatives appear in a wide range of physical systems, from creased thin sheets to materials with sharp stiffness transitions. Accurately modeling these features is essential for simulation but remains challenging for traditional mesh-based methods, which require discontinuity-aligned remeshing -- entangling geometry with simulation and hindering generalization across shape families. Neural fields offer an appealing alternative by encoding basis functions as smooth…

SEED: A Structural Encoder for Embedding-Driven Decoding in Time Series Prediction with LLMs
Fengze Li, Yue Wang, Yangle Liu, Ming Huang, Dou Hong, Jieming Ma
https://arxiv.org/abs/2506.20167

SEED: A Structural Encoder for Embedding-Driven Decoding in Time Series Prediction with LLMs
Multivariate time series forecasting requires models to simultaneously capture variable-wise structural dependencies and generalize across diverse tasks. While structural encoders are effective in modeling feature interactions, they lack the capacity to support semantic-level reasoning or task adaptation. Conversely, large language models (LLMs) possess strong generalization capabilities but remain incompatible with raw time series inputs. This gap limits the development of unified, transferabl…

Towards Mitigation of Hallucination for LLM-empowered Agents: Progressive Generalization Bound Exploration and Watchdog Monitor
Siyuan Liu, Wenjing Liu, Zhiwei Xu, Xin Wang, Bo Chen, Tao Li
https://arxiv.org/abs/2507.15903

Towards Mitigation of Hallucination for LLM-empowered Agents: Progressive Generalization Bound Exploration and Watchdog Monitor
Empowered by large language models (LLMs), intelligent agents have become a popular paradigm for interacting with open environments to facilitate AI deployment. However, hallucinations generated by LLMs-where outputs are inconsistent with facts-pose a significant challenge, undermining the credibility of intelligent agents. Only if hallucinations can be mitigated, the intelligent agents can be used in real-world without any catastrophic risk. Therefore, effective detection and mitigation of hal…

A Topological Perspective on the Birch and Swinnerton Dyer Conjectures
Maisara Shoeib
https://arxiv.org/abs/2505.19796 https://arxiv.…

A Topological Perspective on the Birch and Swinnerton Dyer Conjectures
This paper presents a topological framework for investigating the Birch and Swinnerton Dyer conjecture through four dimensional embeddings of elliptic curves. We propose a correspondence between the algebraic rank of an elliptic curve and the number of topologically independent loops in its embedding, which appears to be related to the order of vanishing of its L function at s=1. Our computational function F new and its generalization F_(m,s) provide methods for examining this relationship thro…

Parallels Between VLA Model Post-Training and Human Motor Learning: Progress, Challenges, and Trends
Tian-Yu Xiang, Ao-Qun Jin, Xiao-Hu Zhou, Mei-Jiang Gui, Xiao-Liang Xie, Shi-Qi Liu, Shuang-Yi Wang, Sheng-Bin Duan, Fu-Chao Xie, Wen-Kai Wang, Si-Cheng Wang, Ling-Yun Li, Tian Tu, Zeng-Guang Hou
https://arxiv.org/abs/2506.20966…

Parallels Between VLA Model Post-Training and Human Motor Learning: Progress, Challenges, and Trends
Vision-language-action (VLA) models extend vision-language models (VLM) by integrating action generation modules for robotic manipulation. Leveraging strengths of VLM in vision perception and instruction understanding, VLA models exhibit promising generalization across diverse manipulation tasks. However, applications demanding high precision and accuracy reveal performance gaps without further adaptation. Evidence from multiple domains highlights the critical role of post-training to align fou…

On a Generalization of the Christoffel Tree: Epichristoffel Trees
Abhishek Krishnamoorthy (Madras Christian College), Robinson Thamburaj (Madras Christian College), Durairaj Gnanaraj Thomas (Madras Christian College)
https://arxiv.org/abs/2507.15313

On a Generalization of the Christoffel Tree: Epichristoffel Trees
Sturmian words form a family of one-sided infinite words over a binary alphabet that are obtained as a discretization of a line with an irrational slope starting from the origin. A finite version of this class of words called Christoffel words has been extensively studied for their interesting properties. It is a class of words that has a geometric and an algebraic definition, making it an intriguing topic of study for many mathematicians. Recently, a generalization of Christoffel words for an …

Complete Minimal Form Factors for Irrelevant Deformations of Integrable Quantum Field Theory
Fabio Sailis, Olalla A. Castro-Alvaredo, Stefano Negro
https://arxiv.org/abs/2506.20517

Complete Minimal Form Factors for Irrelevant Deformations of Integrable Quantum Field Theory
In this paper, we present a method to compute the minimal form factors of diagonal integrable field theories perturbed by generalized $T \bar{T}$ perturbations. Building on existing results by the same authors, these MFFs are constructed in such a way as not to allow for any free parameters, an issue that plagued previous solutions. The MFFs are derived from a generalization of the standard integral representation which has been used since the birth of the form factor bootstrap program. Their a…

S-r-$ideals in Commutative Rings
Abuzer G\"und\"uz, Osama A. Naji, Mehmet \"Ozen
https://arxiv.org/abs/2505.19198 https://

S-r-$ideals in Commutative Rings
This article introduces the notion of $S-r-$ideals in commutative ring $H$ as a generalization of $ r-$ideals. Let $S$ be a multiplicatively closed subset and $A$ an ideal of $H$ with $S\cap A= \emptyset$. $A$ is called $S-r-ideal$ if $ \exists s\in S$ such that for $w,z\in H,\; wz\in A$ and $Ann(w)=0$, then $sz\in A.$ Basic properties of $S-r-$ideals are given. It is shown that an $r-$ideal is always an $S-r-$ideal, and the converse is true under some conditions. Various characterizations of $…

Replaced article(s) found for cs.NE. https://arxiv.org/list/cs.NE/new
[1/1]:
- Enhancing Generalization of Spiking Neural Networks Through Temporal Regularization
Boxuan Zhang, Zhen Xu, Kuan Tao

A Further Generalization of the Gale-Nikaido-Debreu Market Equilibrium Theorem
Ranjit Vohra
https://arxiv.org/abs/2507.17669 https://arxiv.org/pdf/2507.176…

A Further Generalization of the Gale-Nikaido-Debreu Market Equilibrium Theorem
We extend the important generalizations by Yannelis [25] and Cornet et al [7] of the classical result of Gale, Nikaido and Debreu (the "GND theorem") regarding existence of market equilibrium, by broadening the applicability of their results, which apply only to economies with commodity space that can be modeled by a locally convex Hausdorff space, to the wider class of economies with commodity spaces describable by any Hausdorff topological vector space with algebraic dual that separates point…

Diverse and Adaptive Behavior Curriculum for Autonomous Driving: A Student-Teacher Framework with Multi-Agent RL
Ahmed Abouelazm, Johannes Ratz, Philip Sch\"orner, J. Marius Z\"ollner
https://arxiv.org/abs/2507.19146

Diverse and Adaptive Behavior Curriculum for Autonomous Driving: A Student-Teacher Framework with Multi-Agent RL
Autonomous driving faces challenges in navigating complex real-world traffic, requiring safe handling of both common and critical scenarios. Reinforcement learning (RL), a prominent method in end-to-end driving, enables agents to learn through trial and error in simulation. However, RL training often relies on rule-based traffic scenarios, limiting generalization. Additionally, current scenario generation methods focus heavily on critical scenarios, neglecting a balance with routine driving beh…

Indiscernible extraction at small large cardinals from a higher-arity stability notion
James E. Hanson
https://arxiv.org/abs/2506.19147 https://

Indiscernible extraction at small large cardinals from a higher-arity stability notion
We introduce a higher-arity stability notion defined in terms of $k$-splitting, a higher-arity generalization of splitting. We show that theories with bounded $k$-splitting have improved indiscernible extraction at $k$-ineffable cardinals, and we give a non-trivial example of a theory with bounded $k$-splitting but unbounded $(k-1)$-splitting for each odd $k > 1$. We also show that bounded $k$-splitting implies $\mathrm{NFOP}_k$, a higher arity stability notion introduced by Terry and Wolf. We …

Ginsparg-Wilson Hamiltonians with Improved Chiral Symmetry
Hersh Singh
https://arxiv.org/abs/2505.20419 https://arxiv.org/pdf/2505.20…

Ginsparg-Wilson Hamiltonians with Improved Chiral Symmetry
We construct a family of Ginsparg-Wilson Hamiltonians with improved chiral properties, starting from a construction of Creutz-Horvath-Neuberger that provides a doubler-free Hamiltonian lattice regularization for Dirac fermions in even spacetime dimensions. We use a higher-order generalization of the Ginsparg-Wilson relation due to Fujikawa, which yields an order-$k$ Hamiltonian overlap operator for each integer $k \geq 0$, with an exactly conserved but nonquantized chiral charge that becomes qu…

High-Dimensional Data Classification in Concentric Coordinates
Alice Williams, Boris Kovalerchuk
https://arxiv.org/abs/2507.18450 https://arxiv.org/pdf/250…

High-Dimensional Data Classification in Concentric Coordinates
The visualization of multi-dimensional data with interpretable methods remains limited by capabilities for both high-dimensional lossless visualizations that do not suffer from occlusion and that are computationally capable by parameterized visualization. This paper proposes a low to high dimensional data supporting framework using lossless Concentric Coordinates that are a more compact generalization of Parallel Coordinates along with former Circular Coordinates. These are forms of the General…

Extended $\mathcal{O}$-operators, Novikov Yang-Baxter equations and post-Novikov algebras
Jianfeng Yu, Yanyong Hong
https://arxiv.org/abs/2505.20735 https:…

Extended $\mathcal{O}$-operators, Novikov Yang-Baxter equations and post-Novikov algebras
In this paper, we introduce the definition of extended $\mathcal{O}$-operators on a Novikov algebra $(A,\circ)$ associated to an $A$-bimodule Novikov algebra which is a generalization of the definition of $\mathcal{O}$-operators and show that there are new Novikov algebra structures on the $A$-bimodule Novikov algebra obtained from extended $\mathcal{O}$-operators. We also introduce the definition of post-Novikov algebras and show that there is a close relationship between post-Novikov algebras…

IPFormer: Visual 3D Panoptic Scene Completion with Context-Adaptive Instance Proposals
Markus Gross, Aya Fahmy, Danit Niwattananan, Dominik Muhle, Rui Song, Daniel Cremers, Henri Mee{\ss}
https://arxiv.org/abs/2506.20671

IPFormer: Visual 3D Panoptic Scene Completion with Context-Adaptive Instance Proposals
Semantic Scene Completion (SSC) has emerged as a pivotal approach for jointly learning scene geometry and semantics, enabling downstream applications such as navigation in mobile robotics. The recent generalization to Panoptic Scene Completion (PSC) advances the SSC domain by integrating instance-level information, thereby enhancing object-level sensitivity in scene understanding. While PSC was introduced using LiDAR modality, methods based on camera images remain largely unexplored. Moreover, …

GPU-Accelerated Interpretable Generalization for Rapid Cyberattack Detection and Forensics
Shu-Ting Huang, Wen-Cheng Chung, Hao-Ting Pai
https://arxiv.org/abs/2507.14222

GPU-Accelerated Interpretable Generalization for Rapid Cyberattack Detection and Forensics
The Interpretable Generalization (IG) mechanism recently published in IEEE Transactions on Information Forensics and Security delivers state-of-the-art, evidence-based intrusion detection by discovering coherent normal and attack patterns through exhaustive intersect-and-subset operations-yet its cubic-time complexity and large intermediate bitsets render full-scale datasets impractical on CPUs. We present IG-GPU, a PyTorch re-architecture that offloads all pairwise intersections and subset eva…

Coupled Entropy: A Goldilocks Generalization?
Kenric P. Nelson
https://arxiv.org/abs/2506.17229 https://arxiv.org/pdf/2506.17229

Coupled Entropy: A Goldilocks Generalization?
Nonextensive Statistical Mechanics (NSM) has developed into a powerful toolset for modeling and analyzing complex systems. Despite its many successes, a puzzle arose early in its development. The constraints on the Tsallis entropy are in the form of an escort distribution with elements proportional to $p_i^q$, but this same factor within the Tsallis entropy function is not normalized. This led to consideration of the Normalized Tsallis Entropy (NTE); however, the normalization proved to make th…

A conjectural generalization of the Brian\c{c}on-Iarrobino Conjecture
Alexia Ascott, Fatemeh Rezaee, Zhichen Zhou
https://arxiv.org/abs/2506.17704 https://…

A conjectural generalization of the Briançon-Iarrobino Conjecture
We conjecturally generalize the Conjecture by Briançon and Iarrobino on the most singular points of the Hilbert scheme of a tetrahedral number of points in $\BA^3$ by introducing the notion of \textit{locally-maximal singularity}. The formulation of this conjecture is based on the shape of the ideals via a picturesque pattern. It also generalizes the conjectural necessary condition for the most singular points, suggested by the second-named author in \cite{Rezaee-23-Conjectures}, restricted to…

On an Abstraction of Lyapunov and Lagrange Stability
Michelangelo Bin, David Angeli
https://arxiv.org/abs/2507.15047 https://arxiv.or…

On an Abstraction of Lyapunov and Lagrange Stability
This paper studies a set-theoretic generalization of Lyapunov and Lagrange stability for abstract systems described by set-valued maps. Lyapunov stability is characterized as the property of inversely mapping filters to filters, Lagrange stability as that of mapping ideals to ideals. These abstract definitions unveil a deep duality between the two stability notions, enable a definition of global stability for abstract systems, and yield an agile generalization of the stability theorems for basi…

Optical Billiards
Sylvie Oliffson Kamphorst, S\^onia Pinto-de-Carvalho, C\'assio Henrique Vieira Morais
https://arxiv.org/abs/2507.18367 https://arxiv.…

Optical Billiards
This paper investigates the dynamics of optical billiards, a generalization of classic billiards, where light rays travel within a refractive medium and reflect elastically at the boundary. Inspired by studies of acoustic modes in rapidly rotating stars, optical billiards are approached using Riemannian geometry, treating the refractive index as a metric. The discrete dynamics is explored through geodesics, curvature properties, highlighting similarities and differences with classical billiards…

A q-Supercongruence Motivated by Higher-Order Generalized Lehmer-Euler Numbers
Wei-Wei Qi
https://arxiv.org/abs/2507.16825 https://ar…

A q-Supercongruence Motivated by Higher-Order Generalized Lehmer-Euler Numbers
Certain generalization of Euler numbers was defined in 1935 by Lehmer using cubic roots of unity, as a natural generalization of Bernoulli and Euler numbers. In this paper, we define a new polynomial related to the higher-order generalized Lehmer-Euler numbers and determine its a q-supercongruence.

An adversary bound for quantum signal processing
Lorenzo Laneve
https://arxiv.org/abs/2506.20484 https://arxiv.org/pdf/2506.20484

An adversary bound for quantum signal processing
Quantum signal processing (QSP) and quantum singular value transformation (QSVT), have emerged as unifying frameworks in the context of quantum algorithm design. These techniques allow to carry out efficient polynomial transformations of matrices block-encoded in unitaries, involving a single ancilla qubit. Recent efforts try to extend QSP to the multivariate setting (M-QSP), where multiple matrices are transformed simultaneously. However, this generalization faces problems not encountered in t…

This https://arxiv.org/abs/2408.05481 has been replaced.
link: https://scholar.google.com/scholar?q=a

Covariant Electromagnetism in Past-Light-Cone Formalism
We present a manifestly covariant formulation of relativistic electromagnetism, focusing on the computation of electromagnetic fields from moving charges in a manifestly Lorentz-covariant manner. The electromagnetic field at a given spacetime point is determined by the motion of point charges at the intersection of their worldlines with the past light cone of the spacetime point, ensuring causal consistency. This formalism provides a manifestly covariant generalization of the Liénard-Wiechert …

Multiprocessor Scheduling with Memory Constraints: Fundamental Properties and Finding Optimal Solutions
P\'al Andr\'as Papp, Toni B\"ohnlein, A. N. Yzelman
https://arxiv.org/abs/2507.17411

Multiprocessor Scheduling with Memory Constraints: Fundamental Properties and Finding Optimal Solutions
We study the problem of scheduling a general computational DAG on multiple processors in a 2-level memory hierarchy. This setting is a natural generalization of several prominent models in the literature, and it simultaneously captures workload balancing, communication, and data movement due to cache size limitations. We first analyze the fundamental properties of this problem from a theoretical perspective, such as its computational complexity. We also prove that optimizing parallelization and…

Continuous sample space reducing stochastic process
Rahul Chhimpa, Avinash Chand Yadav\
https://arxiv.org/abs/2507.18086 https://arxiv.org/pdf/2507.18086…

Continuous sample space reducing stochastic process
We propose a simple model for sample space reducing (SSR) stochastic process, where the dynamical variable denoting the size of the state space is continuous. In general, one can view the model as a multiplicative stochastic process, with a constraint that the size of the state space cannot be smaller than a visibility parameter $ε$. We study the survival time statistics that reveal a subtle difference from the discrete version of the process. A straightforward generalization can explain the n…

Deep Learning for Blood-Brain Barrier Permeability Prediction
Zihan Yang, Haipeng Gong
https://arxiv.org/abs/2507.18557 https://arxiv.org/pdf/2507.18557

Deep Learning for Blood-Brain Barrier Permeability Prediction
Predicting whether a molecule can cross the blood-brain barrier (BBB) is a key step in early-stage neuropharmaceutical development, directly influencing both research efficiency and success rates in drug discovery. Traditional empirical methods based on physicochemical properties are prone to systematic misjudgements due to their reliance on static rules. Early machine learning models, although data-driven, often suffer from limited capacity, poor generalization, and insufficient interpretabili…

A New Generalization of the Liouville-Jacobi Identity
Lubomir Markov
https://arxiv.org/abs/2507.14927 https://arxiv.org/pdf/2507.14927

A New Generalization of the Liouville-Jacobi Identity
A generalized Liouville-Jacobi Identity is proved for the determinant $\det{X(t)}$ of a solution $X(t)$ to the linear nonhomogeneous first-order matrix differential equation with left- and right-coefficient matrices $\ \frac{\rm d}{{\rm d}t} X(t) + A(t)X(t) + X(t)B(t)= F(t), \ X(t_0)=X_0.$

On Fermat's Last Theorem over the $\mathbb{Z}_3$-extension of $\mathbb{Q}$ and other fields
Luis Dieulefait, Franco Golfieri Madriaga
https://arxiv.org/abs/2507.16883

On Fermat's Last Theorem over the $\mathbb{Z}_3$-extension of $\mathbb{Q}$ and other fields
The main result of the present article is a proof of Fermat's Last Theorem for sufficiently large prime exponents $p$ with $p \equiv 2 \pmod{3}$ over certain number fields. A particular case of these fields are the maximal real subfields of the cyclotomic extensions $\mathbb{Q}(ζ_{3^n})$ for every $n$. Our strategy consists in combining the modular method with a generalization of an arithmetic result of Pomey to these fields.

$\mathbb{G}$-uniform weighted K-stability for models on klt varieties
Jiyuan Han, Yaxiong Liu
https://arxiv.org/abs/2506.18039 https://

$\mathbb{G}$-uniform weighted K-stability for models on klt varieties
In this paper, we make a generalization of the results in \cite{Li22a} to the singular and weighted setting. In particular, we show that on a polarized projective klt variety, the $\mathbb{G}$-uniform weighted K-stability for models implies the $\mathbb{G}$-coercivity of the weighted Mabuchi functional. In the toric case, we further show that the $(\mathbb{C}^{\times})^n$-uniform $(\mathrm{v},\mathrm{w}\cdot\ell_{\mathrm{ext}})$-weighted K-stability is preserved when perturbing the polarization…

A Compact Cycle Formulation for the Multiperiodic Event Scheduling Problem
Rolf Nelson van Lieshout, Niels Lindner
https://arxiv.org/abs/2507.17566 https://

A Compact Cycle Formulation for the Multiperiodic Event Scheduling Problem
The Periodic Event Scheduling Problem (PESP) is a fundamental model in periodic timetabling for public transport systems, assuming a common period across all events. However, real-world networks often feature heterogeneous service frequencies. This paper studies the Multiperiodic Event Scheduling Problem (MPESP), a generalization of PESP that allows each event to recur at its own individual period. While more expressive, MPESP presents new modeling challenges due to the loss of a global period.…

Local fibrations and morphisms of relative toposes
L\'eo Bartoli, Olivia Caramello
https://arxiv.org/abs/2507.14379 https://arxiv…

Local fibrations and morphisms of relative toposes
We introduce the notion of local fibration, a generalization of the notion of fibration which takes into account the presence of Grothendieck topologies on the two categories, and show that the classical results about fibrations lift to this more general setting. As an application of this notion, we obtain a characterization of the functors between relative sites that induce a morphism between the corresponding relative toposes: these are exactly the morphisms of sites which are morphisms of lo…

From Sharpness to Better Generalization for Speech Deepfake Detection
Wen Huang, Xuechen Liu, Xin Wang, Junichi Yamagishi, Yanmin Qian
https://arxiv.org/abs/2506.11532

From Sharpness to Better Generalization for Speech Deepfake Detection
Generalization remains a critical challenge in speech deepfake detection (SDD). While various approaches aim to improve robustness, generalization is typically assessed through performance metrics like equal error rate without a theoretical framework to explain model performance. This work investigates sharpness as a theoretical proxy for generalization in SDD. We analyze how sharpness responds to domain shifts and find it increases in unseen conditions, indicating higher model sensitivity. Bas…

Revisiting Compositional Generalization Capability of Large Language Models Considering Instruction Following Ability
Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe
https://arxiv.org/abs/2506.15629

Revisiting Compositional Generalization Capability of Large Language Models Considering Instruction Following Ability
In generative commonsense reasoning tasks such as CommonGen, generative large language models (LLMs) compose sentences that include all given concepts. However, when focusing on instruction-following capabilities, if a prompt specifies a concept order, LLMs must generate sentences that adhere to the specified order. To address this, we propose Ordered CommonGen, a benchmark designed to evaluate the compositional generalization and instruction-following abilities of LLMs. This benchmark measures…

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[5/5]:
- Omni-Thinker: Scaling Cross-Domain Generalization in LLMs via Multi-Task RL with Hybrid Rewards
Li, Zhou, Kazemi, Sun, Ghaddar, Alomrani, Ma, Luo, Li, Wen, Hao, Coates, Zhang

A generalization of Rao's theorem to graded $R$-subalgebras of $R[t]$
Diksha Garg, Anjan Gupta
https://arxiv.org/abs/2506.18836 https://

A generalization of Rao's theorem to graded $R$-subalgebras of $R[t]$
Let $R$ be a Noetherian local ring of Krull dimension $d$ such that $(d!)R = R$, and let $A$ be a graded $R$-subalgebra of the polynomial algebra $R[t]$. We prove that every unimodular row of length $d + 1$ over $A$ can be completed to an invertible matrix. This is a generalization of a classical result by Rao, who proved that in the same setting, every unimodular row of length $d + 1$ over $R[t]$ admits a completion to an invertible matrix.

ARSAR-Net: Adaptively Regularized SAR Imaging Network via Non-matrix-inversion ADMM
Shiping Fu, Yufan Chen, Zhe Zhang, Xiaolan Qiu, Qixiang Ye
https://arxiv.org/abs/2506.18324

ARSAR-Net: Adaptively Regularized SAR Imaging Network via Non-matrix-inversion ADMM
Deep unfolding networks have constituted an emerging method for synthetic aperture radar (SAR) imaging. However, baseline unfolding networks, when unfolded from the alternating direction method of multipliers (ADMM), lack generalization capability across scenes as the regularizers are empirically designed. In this study, we introduce a learnable regularizer to the unfolding network and create an adaptively regularized SAR imaging network (ARSAR-Net), which pursues generalization capability acro…

Well-Founded Coalgebras Meet K\"onig's Lemma
Henning Urbat, Thorsten Wi{\ss}mann
https://arxiv.org/abs/2507.18539 https://arxiv.org/pdf/2507.18539…

Well-Founded Coalgebras Meet König's Lemma
König's lemma is a fundamental result about trees with countless applications in mathematics and computer science. In contrapositive form, it states that if a tree is finitely branching and well-founded (i.e. has no infinite paths), then it is finite. We present a coalgebraic version of König's lemma featuring two dimensions of generalization: from finitely branching trees to coalgebras for a finitary endofunctor H, and from the base category of sets to a locally finitely presentable category…

Multi-Granular Discretization for Interpretable Generalization in Precise Cyberattack Identification
Wen-Cheng Chung, Shu-Ting Huang, Hao-Ting Pai
https://arxiv.org/abs/2507.14223

Multi-Granular Discretization for Interpretable Generalization in Precise Cyberattack Identification
Explainable intrusion detection systems (IDS) are now recognized as essential for mission-critical networks, yet most "XAI" pipelines still bolt an approximate explainer onto an opaque classifier, leaving analysts with partial and sometimes misleading insights. The Interpretable Generalization (IG) mechanism, published in IEEE Transactions on Information Forensics and Security, eliminates that bottleneck by learning coherent patterns - feature combinations unique to benign or malicious traffic …

Information-Theoretic Generalization Bounds of Replay-based Continual Learning
Wen Wen, Tieliang Gong, Yunjiao Zhang, Zeyu Gao, Weizhan Zhang, Yong-Jin Liu
https://arxiv.org/abs/2507.12043

Information-Theoretic Generalization Bounds of Replay-based Continual Learning
Continual learning (CL) has emerged as a dominant paradigm for acquiring knowledge from sequential tasks while avoiding catastrophic forgetting. Although many CL methods have been proposed to show impressive empirical performance, the theoretical understanding of their generalization behavior remains limited, particularly for replay-based approaches. In this paper, we establish a unified theoretical framework for replay-based CL, deriving a series of information-theoretic bounds that explicitly…

A Generalization of the B\"{o}ttcher-Wenzel inequality for three rectangular matrices
Motoyuki Nobori
https://arxiv.org/abs/2506.17365 https://…

A Generalization of the Böttcher-Wenzel inequality for three rectangular matrices
Let $m,n$ be positive integers. For all $m\times n$ complex matrices $A, C$ and an $n\times m$ matrix $B$, we define a generalized commutator as $ABC-CBA$. We estimated the Frobenius norm of it, and finally got the inequality, which is a generalization of the Böttcher-Wenzel inequality. If $n=1$ or $m=1$, then the Frobenius norm of $ABC-CBA$ can be estimated with a tighter upper bound.

A Novel Coded Computing Approach for Distributed Multi-Task Learning
Minquan Cheng, Yongkang Wang, Lingyu Zhang, Youlong Wu
https://arxiv.org/abs/2507.18025 https://

A Novel Coded Computing Approach for Distributed Multi-Task Learning
Distributed multi-task learning (DMTL) effectively improves model generalization performance through the collaborative training of multiple related models. However, in large-scale learning scenarios, communication bottlenecks severely limit practical system performance. In this paper, we investigate the communication bottleneck within a typical DMTL system that employs non-linear global updates. This system involves distributed workers, assisted by a central server, who collaboratively learn di…

Continuous spin superparticle in $4D$, ${\cal N}=1$ curved superspace
I. L. Buchbinder, S. A. Fedoruk
https://arxiv.org/abs/2507.18524 https://arxiv.org/pd…

Continuous spin superparticle in $4D$, ${\cal N}=1$ curved superspace
We present a new particle model that describes the dynamics of a $4D,$ $\mathcal{N}{=}\,1$ continuous spin particle in $AdS_4$ superspace and is a generalization of the continuous-spin superparticle model in flat $4D$, $\mathcal{N}{=}\,1$ superspace proposed in 2506.19709 [hep-th]. The model is described by $4D$, $\mathcal{N}{=}\,1$ superspace coordinates together with commuting spinor additional variables, which are inherent ingredients of continuous spin models. The Lagrangian and the system …

A generalization of Deterministic Finite Automata related to discharging
John M. Campbell
https://arxiv.org/abs/2506.14072 https://ar…

A generalization of Deterministic Finite Automata related to discharging
Deterministic Finite Automata (DFAs) are of central importance in automata theory. In view of how state diagrams for DFAs are defined using directed graphs, this leads us to introduce a generalization of DFAs related to a method widely used in graph theory referred to as the discharging method. Given a DFA $(Q, Σ, δ, q_{0}, F)$, the transition function $δ\colon Q \times Σ\to Q$ determines a directed path in the corresponding state diagram based on an input string $a_{1} a_{2} \cdots a_{n}$ …

U-Net Based Healthy 3D Brain Tissue Inpainting
Juexin Zhang, Ying Weng, Ke Chen
https://arxiv.org/abs/2507.18126 https://arxiv.org/pdf/2507.18126

U-Net Based Healthy 3D Brain Tissue Inpainting
This paper introduces a novel approach to synthesize healthy 3D brain tissue from masked input images, specifically focusing on the task of 'ASNR-MICCAI BraTS Local Synthesis of Tissue via Inpainting'. Our proposed method employs a U-Net-based architecture, which is designed to effectively reconstruct the missing or corrupted regions of brain MRI scans. To enhance our model's generalization capabilities and robustness, we implement a comprehensive data augmentation strategy that involves random…

On recognizing graphs representing Persistent Perfect Phylogenies
Paola Bonizzoni, Gianluca Della Vedova, Mauricio Soto Gomez, Gabriella Trucco
https://arxiv.org/abs/2507.18281 …

On recognizing graphs representing Persistent Perfect Phylogenies
The Persistent Perfect phylogeny, also known as Dollo-1, has been introduced as a generalization of the well-known perfect phylogenetic model for binary characters to deal with the potential loss of characters. The problem of deciding the existence of a Persistent Perfect phylogeny can be reduced to the one of recognizing a class of bipartite graphs whose nodes are species and characters. Thus an interesting question is solving directly the problem of recognizing such graphs. We present a polyn…

Projective length, phantom extensions, and the structure of flat modules
Matteo Casarosa, Martino Lupini
https://arxiv.org/abs/2506.17982 https://

Projective length, phantom extensions, and the structure of flat modules
We consider the natural generalization of the notion of the order of a phantom map from the topological setting to triangulated categories. When applied to the derived category of the category of countable flat modules over a countable Dedekind domain, this yields a notion of\emph{\ phantom extension} of order $α<ω_{1}$. We provide a complexity-theoretic characterization of the module $\mathrm{Ph}% ^{α}\mathrm{Ext}\left( C,A\right) $ of phantom extensions of order $α$ with respect to the st…

Identifying Prompted Artist Names from Generated Images
Grace Su, Sheng-Yu Wang, Aaron Hertzmann, Eli Shechtman, Jun-Yan Zhu, Richard Zhang
https://arxiv.org/abs/2507.18633 http…

Identifying Prompted Artist Names from Generated Images
A common and controversial use of text-to-image models is to generate pictures by explicitly naming artists, such as "in the style of Greg Rutkowski". We introduce a benchmark for prompted-artist recognition: predicting which artist names were invoked in the prompt from the image alone. The dataset contains 1.95M images covering 110 artists and spans four generalization settings: held-out artists, increasing prompt complexity, multiple-artist prompts, and different text-to-image models. We eval…

Label Unification for Cross-Dataset Generalization in Cybersecurity NER
Maciej Jalocha, Johan Hausted Schmidt, William Michelseen
https://arxiv.org/abs/2507.13870

Label Unification for Cross-Dataset Generalization in Cybersecurity NER
The field of cybersecurity NER lacks standardized labels, making it challenging to combine datasets. We investigate label unification across four cybersecurity datasets to increase data resource usability. We perform a coarse-grained label unification and conduct pairwise cross-dataset evaluations using BiLSTM models. Qualitative analysis of predictions reveals errors, limitations, and dataset differences. To address unification limitations, we propose alternative architectures including a mult…

A Jarque-Bera test for skew normal data
Diam Ba, Gorgui Gning, Gandasor Bonyiri Onesiphore Da, Oumar Foly Sow, Gane Samb Lo
https://arxiv.org/abs/2507.18032 https://

A Jarque-Bera test for skew normal data
The skew normal law has been introduced in Azzalin (1985) as an alternative to adjusting asymmetric data that share important patterns with the normal law. It has been extensively studied. However, there is so much to do in order to catch the diversity and the richness of the investigation of its normal counterpart. The General Jarque-Berra Test (GJBT) has been devised by Lo et al. (2015), Da et al. (2023) for arbitrary laws with at least finite first eight moments, as a generalization of the J…

Numerical solution of quantum Landau-Lifshitz-Gilbert equation
Vahid Azimi-Mousolou, Davoud Mirzaei
https://arxiv.org/abs/2506.19594 https://

Numerical solution of quantum Landau-Lifshitz-Gilbert equation
The classical Landau-Lifshitz-Gilbert (LLG) equation has long served as a cornerstone for modeling magnetization dynamics in magnetic systems, yet its classical nature limits its applicability to inherently quantum phenomena such as entanglement and nonlocal correlations. Inspired by the need to incorporate quantum effects into spin dynamics, recently a quantum generalization of the LLG equation is proposed [Phys. Rev. Lett. 133, 266704 (2024)] which captures essential quantum behavior in many-…

Generalization Error Analysis for Attack-Free and Byzantine-Resilient Decentralized Learning with Data Heterogeneity
Haoxiang Ye, Tao Sun, Qing Ling
https://arxiv.org/abs/2506.09438

Generalization Error Analysis for Attack-Free and Byzantine-Resilient Decentralized Learning with Data Heterogeneity
Decentralized learning, which facilitates joint model training across geographically scattered agents, has gained significant attention in the field of signal and information processing in recent years. While the optimization errors of decentralized learning algorithms have been extensively studied, their generalization errors remain relatively under-explored. As the generalization errors reflect the scalability of trained models on unseen data and are crucial in determining the performance of …

Factorization in independent sums of Haar system Hardy spaces
Konstantinos Konstantos, Thomas Speckhofer
https://arxiv.org/abs/2507.18600 https://arxiv.org…

Factorization in independent sums of Haar system Hardy spaces
We introduce a generalization of the Bourgain-Rosenthal-Schechtman $R_ω^p$ space: Let $Y$ be a Haar system Hardy space, i.e., a separable rearrangement-invariant function space on the unit interval or an associated Hardy space defined via the square function (such as dyadic $H^1$). Then we define $Y_ω$ as the closed linear span in $Y$ of independent distributional copies of the spaces $Y_n$ of dyadic step functions at scale $2^{-n}$. Combining finite-dimensional and infinite-dimensional techn…

De Rham-Higgs comparison for mixed Hodge modules in positive characteristic
Zhang Zebao
https://arxiv.org/abs/2507.15175 https://arxi…

De Rham-Higgs comparison for mixed Hodge modules in positive characteristic
Building on the nonabelian Hodge theory in positive characteristic developed by Ogus, Vologodsky, and Schepler, we propose a generalization of the decomposition theorem of Deligne and Illusie from the perspective of mixed Hodge modules. This generalization is verified in certain special cases by extending the method of Sheng and the author, leading to several interesting byproducts.

Steering Conceptual Bias via Transformer Latent-Subspace Activation
Vansh Sharma, Venkat Raman
https://arxiv.org/abs/2506.18887 https://

Steering Conceptual Bias via Transformer Latent-Subspace Activation
This work examines whether activating latent subspaces in language models (LLMs) can steer scientific code generation toward a specific programming language. Five causal LLMs were first evaluated on scientific coding prompts to quantify their baseline bias among four programming languages. A static neuron-attribution method, perturbing the highest activated MLP weight for a C++ or CPP token, proved brittle and exhibited limited generalization across prompt styles and model scales. To address th…

Subcoloring of (Unit) Disk Graphs
Malory Marin, R\'emi Watrigant
https://arxiv.org/abs/2506.19452 https://arxiv.org/pdf/2506.1945…

Subcoloring of (Unit) Disk Graphs
A subcoloring of a graph is a partition of its vertex set into subsets (called colors), each inducing a disjoint union of cliques. It is a natural generalization of the classical proper coloring, in which each color must instead induce an independent set. Similarly to proper coloring, we define the subchromatic number of a graph as the minimum integer k such that it admits a subcoloring with k colors, and the corresponding problem k-Subcoloring which asks whether a graph has subchromatic number…

Cubes-difference factor absorbing ideals of a commutative ring
Faranak Farshadifar
https://arxiv.org/abs/2506.19876 https://arxiv.org…

Cubes-difference factor absorbing ideals of a commutative ring
Let $R$ be a commutative ring with. The purpose of this paper is to introduce and investigate cubes-difference factor absorbing ideals of R as a generalization of prime ideals.

AQuilt: Weaving Logic and Self-Inspection into Low-Cost, High-Relevance Data Synthesis for Specialist LLMs
Xiaopeng Ke, Hexuan Deng, Xuebo Liu, Jun Rao, Zhenxi Song, Jun Yu, Min Zhang
https://arxiv.org/abs/2507.18584

AQuilt: Weaving Logic and Self-Inspection into Low-Cost, High-Relevance Data Synthesis for Specialist LLMs
Despite the impressive performance of large language models (LLMs) in general domains, they often underperform in specialized domains. Existing approaches typically rely on data synthesis methods and yield promising results by using unlabeled data to capture domain-specific features. However, these methods either incur high computational costs or suffer from performance limitations, while also demonstrating insufficient generalization across different tasks. To address these challenges, we prop…

ICWLM: A Multi-Task Wireless Large Model via In-Context Learning
Yuxuan Wen, Xiaoming Chen, Maojun Zhang, Zhaoyang Zhang
https://arxiv.org/abs/2507.18167 https://

ICWLM: A Multi-Task Wireless Large Model via In-Context Learning
The rapid evolution of wireless communication technologies, particularly massive multiple-input multiple-output (mMIMO) and millimeter-wave (mmWave), introduces significant network complexity and computational demands. Significant research efforts have been made to improve physical layer performance by resorting to deep learning (DL) methods, which, however, are usually task-specific and struggle with data scarcity and generalization. To address these challenges, we propose a novel In-Context W…

Adaptive Articulated Object Manipulation On The Fly with Foundation Model Reasoning and Part Grounding
Xiaojie Zhang, Yuanfei Wang, Ruihai Wu, Kunqi Xu, Yu Li, Liuyu Xiang, Hao Dong, Zhaofeng He
https://arxiv.org/abs/2507.18276

Adaptive Articulated Object Manipulation On The Fly with Foundation Model Reasoning and Part Grounding
Articulated objects pose diverse manipulation challenges for robots. Since their internal structures are not directly observable, robots must adaptively explore and refine actions to generate successful manipulation trajectories. While existing works have attempted cross-category generalization in adaptive articulated object manipulation, two major challenges persist: (1) the geometric diversity of real-world articulated objects complicates visual perception and understanding, and (2) variation…

Sharp Generalization Bounds for Foundation Models with Asymmetric Randomized Low-Rank Adapters
Anastasis Kratsios, Tin Sum Cheng, Aurelien Lucchi, Haitz S\'aez de Oc\'ariz Borde
https://arxiv.org/abs/2506.14530

Sharp Generalization Bounds for Foundation Models with Asymmetric Randomized Low-Rank Adapters
Low-Rank Adaptation (LoRA) has emerged as a widely adopted parameter-efficient fine-tuning (PEFT) technique for foundation models. Recent work has highlighted an inherent asymmetry in the initialization of LoRA's low-rank factors, which has been present since its inception and was presumably derived experimentally. This paper focuses on providing a comprehensive theoretical characterization of asymmetric LoRA with frozen random factors. First, while existing research provides upper-bound genera…

Confidence Optimization for Probabilistic Encoding
Pengjiu Xia, Yidian Huang, Wenchao Wei, Yuwen Tan
https://arxiv.org/abs/2507.16881 https://

Confidence Optimization for Probabilistic Encoding
Probabilistic encoding introduces Gaussian noise into neural networks, enabling a smooth transition from deterministic to uncertain states and enhancing generalization ability. However, the randomness of Gaussian noise distorts point-based distance measurements in classification tasks. To mitigate this issue, we propose a confidence optimization probabilistic encoding (CPE) method that improves distance reliability and enhances representation learning. Specifically, we refine probabilistic enco…

On the High-Rate FDPC Codes: Construction, Encoding, and a Generalization
Mohsen Moradi, Sheida Rabeti, Hessam Mahdavifar
https://arxiv.org/abs/2506.11345 …

On the High-Rate FDPC Codes: Construction, Encoding, and a Generalization
Recently introduced Fair-Density Parity-Check (FDPC) codes, targeting high-rate applications, offer superior error-correction performance (ECP) compared to 5G Low-Density Parity-Check (LDPC) codes, given the same number of message-passing decoding iterations. In this paper, we present a novel construction method for FDPC codes, introduce a generalization of these codes, and propose a low-complexity encoding algorithm. Numerical results demonstrate the fast convergence of the message-passing dec…

Domain Generalization for Person Re-identification: A Survey Towards Domain-Agnostic Person Matching
Hyeonseo Lee, Juhyun Park, Jihyong Oh, Chanho Eom
https://arxiv.org/abs/2506.12413

Domain Generalization for Person Re-identification: A Survey Towards Domain-Agnostic Person Matching
Person Re-identification (ReID) aims to retrieve images of the same individual captured across non-overlapping camera views, making it a critical component of intelligent surveillance systems. Traditional ReID methods assume that the training and test domains share similar characteristics and primarily focus on learning discriminative features within a given domain. However, they often fail to generalize to unseen domains due to domain shifts caused by variations in viewpoint, background, and l…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[5/11]:
- Memorization to Generalization: Emergence of Diffusion Models from Associative Memory
Bao Pham, Gabriel Raya, Matteo Negri, Mohammed J. Zaki, Luca Ambrogioni, Dmitry Krotov

Moduli Theory of the $r$-Braid Arrangement
Vance Blankers, Emily Clader, Iva Halacheva, Haggai Liu, Dustin Ross
https://arxiv.org/abs/2506.18205 https://…

Moduli Theory of the $r$-Braid Arrangement
We describe a family of hyperplane arrangements depending on a positive integer parameter $r$, which we refer to as the $r$-braid arrangements, and which can be viewed as a generalization of the classical braid arrangement. The wonderful compactification of the braid arrangement (with respect to its minimal building set) is well-known to yield the moduli space $\overline{\mathcal{M}}_{0,n}$, and, in this work, we generalize this result, constructing a moduli space $\overline{\mathcal{M}}^r_{n}$…

Replaced article(s) found for cs.IT. https://arxiv.org/list/cs.IT/new
[1/1]:
- Fast Rate Information-theoretic Bounds on Generalization Errors
Xuetong Wu, Jonathan H. Manton, Uwe Aickelin, Jingge Zhu

SViP: Sequencing Bimanual Visuomotor Policies with Object-Centric Motion Primitives
Yizhou Chen, Hang Xu, Dongjie Yu, Zeqing Zhang, Yi Ren, Jia Pan
https://arxiv.org/abs/2506.18825

SViP: Sequencing Bimanual Visuomotor Policies with Object-Centric Motion Primitives
Imitation learning (IL), particularly when leveraging high-dimensional visual inputs for policy training, has proven intuitive and effective in complex bimanual manipulation tasks. Nonetheless, the generalization capability of visuomotor policies remains limited, especially when small demonstration datasets are available. Accumulated errors in visuomotor policies significantly hinder their ability to complete long-horizon tasks. To address these limitations, we propose SViP, a framework that se…

CMP: A Composable Meta Prompt for SAM-Based Cross-Domain Few-Shot Segmentation
Shuai Chen, Fanman Meng, Chunjin Yang, Haoran Wei, Chenhao Wu, Qingbo Wu, Hongliang Li
https://arxiv.org/abs/2507.16753

CMP: A Composable Meta Prompt for SAM-Based Cross-Domain Few-Shot Segmentation
Cross-Domain Few-Shot Segmentation (CD-FSS) remains challenging due to limited data and domain shifts. Recent foundation models like the Segment Anything Model (SAM) have shown remarkable zero-shot generalization capability in general segmentation tasks, making it a promising solution for few-shot scenarios. However, adapting SAM to CD-FSS faces two critical challenges: reliance on manual prompt and limited cross-domain ability. Therefore, we propose the Composable Meta-Prompt (CMP) framework t…

C3RL: Rethinking the Combination of Channel-independence and Channel-mixing from Representation Learning
Shusen Ma, Yun-Bo Zhao, Yu Kang
https://arxiv.org/abs/2507.17454

C3RL: Rethinking the Combination of Channel-independence and Channel-mixing from Representation Learning
Multivariate time series forecasting has drawn increasing attention due to its practical importance. Existing approaches typically adopt either channel-mixing (CM) or channel-independence (CI) strategies. CM strategy can capture inter-variable dependencies but fails to discern variable-specific temporal patterns. CI strategy improves this aspect but fails to fully exploit cross-variable dependencies like CM. Hybrid strategies based on feature fusion offer limited generalization and interpretabi…

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[5/6]:
- Quantifying artificial intelligence through algorithmic generalization
Takuya Ito, Murray Campbell, Lior Horesh, Tim Klinger, Parikshit Ram

A derived category analogue of the Nakai--Moishezon criterion
Daigo Ito, Noah Olander
https://arxiv.org/abs/2507.17681 https://arxiv.org/pdf/2507.17681

A derived category analogue of the Nakai--Moishezon criterion
We give a complete characterization of the line bundles on a proper variety whose tensor powers generate the derived category, answering a 2010 question of Chris Brav. The condition is analogous to the Nakai--Moishezon criterion and can be stated purely in terms of classical notions of positivity of line bundles. There is also a generalization which works for all Noetherian schemes. We use our criterion to prove basic properties of such line bundles and provide non-trivial examples of them. As …

Fair Domain Generalization: An Information-Theoretic View
Tangzheng Lian, Guanyu Hu, Dimitrios Kollias, Xinyu Yang, Oya Celiktutan
https://arxiv.org/abs/2507.05823

Fair Domain Generalization: An Information-Theoretic View
Domain generalization (DG) and algorithmic fairness are two critical challenges in machine learning. However, most DG methods focus only on minimizing expected risk in the unseen target domain without considering algorithmic fairness. Conversely, fairness methods typically do not account for domain shifts, so the fairness achieved during training may not generalize to unseen test domains. In this work, we bridge these gaps by studying the problem of Fair Domain Generalization (FairDG), which ai…

Impact of Pretraining Word Co-occurrence on Compositional Generalization in Multimodal Models
Helen Qu, Sang Michael Xie
https://arxiv.org/abs/2507.08000 h…

Impact of Pretraining Word Co-occurrence on Compositional Generalization in Multimodal Models
CLIP and large multimodal models (LMMs) have better accuracy on examples involving concepts that are highly represented in the training data. However, the role of concept combinations in the training data on compositional generalization is largely unclear -- for instance, how does accuracy vary when a common object appears in an uncommon pairing with another object? In this paper, we investigate how word co-occurrence statistics in the pretraining dataset (a proxy for co-occurrence of visual co…

Rethinking LLM Training through Information Geometry and Quantum Metrics
Riccardo Di Sipio
https://arxiv.org/abs/2506.15830 https://a…

Rethinking LLM Training through Information Geometry and Quantum Metrics
Optimization in large language models (LLMs) unfolds over high-dimensional parameter spaces with non-Euclidean structure. Information geometry frames this landscape using the Fisher information metric, enabling more principled learning via natural gradient descent. Though often impractical, this geometric lens clarifies phenomena such as sharp minima, generalization, and observed scaling laws. We argue that curvature-aware approaches deepen our understanding of LLM training. Finally, we specula…

From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models
Irving Fang, Juexiao Zhang, Shengbang Tong, Chen Feng
https://arxiv.org/abs/2506.09930

From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models
One promise that Vision-Language-Action (VLA) models hold over traditional imitation learning for robotics is to leverage the broad generalization capabilities of large Vision-Language Models (VLMs) to produce versatile, "generalist" robot policies. However, current evaluations of VLAs remain insufficient. Traditional imitation learning benchmarks are unsuitable due to the lack of language instructions. Emerging benchmarks for VLAs that incorporate language often come with limited evaluation ta…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/9]:
- On the generalization of Tanimoto-type kernels to real valued functions
Sandor Szedmak, Eric Bach