Newest journal articles by subject

Subjects > Mathematics

Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
Abdelrahman Abouelenin;Atabak AshfaqAdam AtkinsonH. AwadallaNguyen BachJianmin BaoAlon BenhaimMartin CaiVishrav ChaudharyCongcong ChenDongdong ChenDongdong ChenJunkun ChenWeizhu ChenYen-Chun ChenYi-Ling ChenQi DaiXiyang DaiRuchao FanMei GaoMingcheng GaoAmit GargAbhishek GoswamiJunheng HaoAmr HendyYuxuan HuXin JinMahmoud KhademiDongwoo KimYoung Jin KimGina LeeJinyu LiYunsheng LiChen LiangXihui LinZeqi LinMeng-Jie LiuYang LiuGilsinia LopezChong LuoPiyush MadanV. MazalovAli MousaviA. NguyenJing PanD. Perez-BeckerJacob PlatinThomas PortetKai QiuBo RenLiliang RenSambuddha RoyNing ShangYelong ShenSaksham SinghalSubhojit SomXiaocheng SongT. SychPraneetha VaddamanuShuohang WangYiming WangZhenghao WangHaibin WuHaoran XuWeijian XuYifan YangZiyi YangDonghan YuI. ZabirJianwen ZhangL. ZhangYunan ZhangXiren Zhou
ArXiv Published 2025/03/03

Summary:

Despite its compact 3.8-billion-parameter size, this experimental version of Phi-4-Mini achieves reasoning performance on par with or surpassing significantly larger models, including DeepSeek-R1-Distill-Qwen-7B and DeepSeek-R1-Distill-Llama-8B.

OpenThoughts: Data Recipes for Reasoning Models
E. Guha;Ryan MartenSedrick Scott KehNegin RaoofG. SmyrnisHritik BansalMarianna NezhurinaJean-Pierre MercatTrung VuZayne SpragueAshima SuvarnaBen FeuerLiangyu ChenZaid KhanEric FrankelSachin GroverCaroline ChoiNiklas MuennighoffShiye SuWanjia ZhaoJohn YangShreyas PimpalgaonkarK. SharmaCharlie Cheng-Jie JiYichuan DengSarah PrattV. RamanujanJon Saad-FalconJeffrey LiAchal DaveAlon AlbalakK. AroraBlake WulfeChinmay HegdeGreg DurrettSewoong OhMohit BansalSaadia GabrielAditya GroverKai-Wei ChangVaishaal ShankarAaron GokaslanMike A. MerrillTatsunori HashimotoYejin ChoiJ. JitsevReinhard HeckelM. SathiamoorthyAlexandros G. DimakisLudwig Schmidt
ArXiv Published 2025/06/04

Summary:

The goal of the OpenThoughts project is to create open-source datasets for training reasoning models and to create the first model trained on public reasoning data to match DeepSeek-R1-Distill-Qwen-7B on standard reasoning benchmarks such as AIME and LiveCodeBench.

Quantum groups and Yang-Baxter equations
A. Isaev
Natural Science Review Published 2025/03/31
SycEval: Evaluating LLM Sycophancy
A.H. Fanous;Jacob GoldbergAnk A. AgarwalJoanna LinAnson Y. ZhouR. DaneshjouOluwasanmi Koyejo
ArXiv Published 2025/02/12

Summary:

A framework to evaluate sycophantic behavior in ChatGPT-4o, Claude-Sonnet, and Gemini-1.5-Pro across AMPS (mathematics) and MedQuad (medical advice) datasets is introduced.

How to Interpret Statistical Models Using marginaleffects for R and Python
Vincent Arel-Bundock;Noah GreiferA. Heiss
J. Stat. Softw.
Olympiad-level formal mathematical reasoning with reinforcement learning
T. Hubert;Rishi S MehtaLaurent SartranMiklós Z HorváthGoran ŽužićEric WieserAja HuangJulian SchrittwieserYannick SchroeckerHussain MasoomOttavia BertolliTom ZahavyAmol MandhaneJessica YungIuliya BeloshapkaBorja IbarzVivek VeeriahLei YuOliver NashPaul LezeauSalvatore MercuriCalle SönneB. MehtaA. DaviesDaniel ZhengF. PedregosaYin LiIngrid von GlehnM. RowlandSamuel AlbanieAmeya VelingkerS. SchmittEdward LockhartEdward HughesH. MichalewskiNicolas SonneratD. HassabisP. KohliDavid Silver
Nature Published 2025/11/12

Summary:

AlphaProof is presented, an AlphaZero-inspired2 agent that learns to find formal proofs through RL by training on millions of auto-formalized problems, and substantially improves state-of-the-art results on historical mathematics competition problems.

My Favorite Things
A. Klein
Mathematics Teacher: Learning and Teaching PK-12 Published 2025/05/01
The Unreasonable Ineffectiveness of the Deeper Layers
Andrey Gromov;Kushal TirumalaHassan ShapourianPaolo GloriosoDaniel A. Roberts
ArXiv Published 2024/03/26

Summary:

This study studies layer pruning via parameter-efficient finetuning methods, specifically quantization and Low Rank Adapters (QLoRA), such that each of the experiments can be performed on a single 40GB A100 GPU.

REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards
Zafir Stojanovski;Oliver StanleyJoe SharrattRichard JonesA. AdefioyeJean KaddourAndreas Köpf
ArXiv Published 2025/05/30

Summary:

The experimental results demonstrate the efficacy of RG in both evaluating and reinforcement learning of reasoning models, and its key innovation is the ability to generate virtually infinite training data with adjustable complexity, unlike most previous reasoning datasets, which are typically fixed.

Problem Posing as a Learning Model to Improve Primary School Students' Mathematics Learning Outcomes in Gayo Lues
Sinar Rahmah;A. H. Lubis
Journal of Indonesian Primary School Published 2024/12/31
Hybrid approaches to optimization and machine learning methods: a systematic literature review
Beatriz Flamia Azevedo;Ana Maria A. C. RochaAna I. Pereira
Machine Learning Published 2024/01/24

Summary:

An extensive systematic and bibliometric literature review on hybrid methods involving optimization and machine learning techniques for clustering and classification aims to identify the potential of methods and algorithms to overcome the difficulties of one or both methodologies when combined.

Practical Efficiency of Muon for Pretraining
Essential AI Darsh J Shah;Anthony M. PollorenoKarl StratosPhilip MonkAdarsh ChaluvarajuAndrew HojelAndrew MaAnil ThomasA. TanwerDarsh J ShahKhoi NguyenKurt SmithMichael CallahanMichael PustMohit ParmarPeter RushtonPlaton MazarakisRitvik KapilaSaurabh SrivastavaSomanshu SinglaT. RomanskiYash VanjaniAshish Vaswani
ArXiv Published 2025/05/04
Recent Advances in Grey Wolf Optimizer, its Versions and Applications: Review
S. Makhadmeh;M. Al-BetarIyad Abu DoushMohammed A. AwadallahSofian KassaymehSeyedali MirjaliliR. A. Zitar
IEEE Access

Summary:

This review delves into the GWO-related research conducted between 2019 and 2022, encompassing over 200 research articles and explores the growth of GWO in terms of publications, citations, and the domains that leverage its potential.

Hyperparameter Tuning in Machine Learning: A Comprehensive Review
Justus A Ilemobayo;O. DurodolaOreoluwa AladeOpeyemi J AwotundeAdewumi T OlanrewajuOlumide Babatope FalanaAdedolapo OgungbireAbraham OsinugaDabira OgunbiyiArk O. IfeanyiIkenna E OdezuligboOluwagbotemi E Edu
Journal of Engineering Research and Reports Published 2024/06/07

Summary:

This review explores the critical role of hyperparameter tuning in ML, detailing its importance, applications, and various optimization techniques, and various tuning methods, including grid search, random search, Bayesian optimization, and meta-learning.

The Leaderboard Illusion
Shivalika Singh;Yiyang NanAlex WangDaniel D'souzaSayash KapoorA. UstunSanmi KoyejoYuntian DengShayne LongpreNoah SmithB. ErmişMarzieh FadaeeSara Hooker
ArXiv Published 2025/04/29

Summary:

This work identifies systematic issues that have resulted in a distorted playing field in Chatbot Arena and offers actionable recommendations to reform the Chatbot Arena's evaluation framework and promote fairer, more transparent benchmarking for the field.

Next Generation Advanced Transceiver Technologies for 6G and Beyond
Changsheng You;Yunlong CaiYuanwei LiuM. Di RenzoTolga M. DumanA. YenerA. Lee Swindlehurst
IEEE Journal on Selected Areas in Communications Published 2024/03/25

Summary:

This tutorial provides an overview of new-field NGAT technology, which shifts from conventional far-field channel models to new near-field channel models, and discusses recent advances in semantic-aware NGAT technologies, which can utilize new metrics for advanced transceiver designs.

The Relationship Between Reasoning and Performance in Large Language Models - o3 (mini) Thinks Harder, Not Longer
Marthe Ballon;A. AlgabaVincent Ginis
ArXiv Published 2025/02/21

Summary:

It is shown that accuracy generally declines as reasoning chains grow across all models and compute settings, even when controlling for difficulty of the questions, and that while o3-mini (h) achieves a marginal accuracy gain over o3-mini (m), it does so by allocating substantially more reasoning tokens across all problems, even the ones that o3-mini (m) can already solve.

Is Model Collapse Inevitable? Breaking the Curse of Recursion by Accumulating Real and Synthetic Data
Matthias Gerstgrasser;Rylan SchaefferApratim DeyRafael RafailovHenry SleightJohn HughesTomasz KorbakRajashree AgrawalDhruv PaiAndrey GromovDaniel A. RobertsDiyi YangD. DonohoOluwasanmi Koyejo
ArXiv Published 2024/04/01

Summary:

It is shown that if data are replaced, the test error increases with the number of model-fitting iterations, but if data instead accumulate, the test error has a finite upper bound independent of the number of iterations, meaning model collapse no longer occurs.

sdmTMB: An R Package for Fast, Flexible, and User-Friendly Generalized Linear Mixed Effects Models with Spatial and Spatiotemporal Random Fields
S. Anderson;E. WardPhilina A. EnglishLewis A. K. BarnettJ. Thorson
bioRxiv Published 2024/07/18

Summary:

The R package sdmTMB is introduced, which extends the flexible interface familiar to users of lme4, glmmTMB, and mgcv to include spatial and spatiotemporal latent GMRFs using an SPDE-(stochastic partial differential equation) based approach and is hoped to help open this useful class of models to a wider field of geostatistical analysts.

Modeling and analysis of the fractional-order epidemic model to investigate mutual influence in HIV/HCV co-infection
Parvaiz Ahmad Naik;B. YeolekarSania QureshiMahesh A. YeolekarA. Madzvamuse
Nonlinear Dynamics Published 2024/05/13

Summary:

A memory trace (MT) procedure is incorporated in this paper that captures and amalgamates the historical dynamics of the system to evoke the memory effect in detail and claims the existence of memory effects of fractional-order derivatives.

Project-based learning as a catalyst for 21st-Century skills and student engagement in the math classroom
N. Rehman;Xiao HuangA. MahmoodMohammed A. M. AlGerafiSaima Javed
Heliyon Published 2024/11/01
DsDm: Model-Aware Dataset Selection with Datamodels
Logan Engstrom;Axel FeldmannA. Ma̧dry
ArXiv Published 2024/01/23

Summary:

This framework frames dataset selection as an optimization problem that the learning process uses train datapoints to predict on the target tasks, and instead models explicitly how the learning process uses train datapoints to predict on the target tasks.

Stop Regressing: Training Value Functions via Classification for Scalable Deep RL
Jesse Farebrother;Jordi OrbayQ. VuongAdrien Ali TaigaYevgen ChebotarTed XiaoA. IrpanSergey LevinePablo Samuel CastroAleksandra FaustAviral KumarRishabh Agarwal
Published 2024/03/06

Summary:

It is argued that a simple shift to training value functions with categorical cross-entropy can yield substantial improvements in the scalability of deep RL at little-to-no cost.

Artificial Intelligence in Education: Mathematics Teachers’ Perspectives, Practices and Challenges
Y. Wardat;Mohammad A. TashtoushRommel AlaliS. Saleh
Iraqi Journal For Computer Science and Mathematics Published 2024/01/03

Summary:

Investigation of mathematics teachers’ perceptions of implemented AI systems and applications in Abu Dhabi Emirate schools revealed that AI could be used as an educational tool to facilitate teaching and develop students’ performance by including AI systems and applications in the curricula.

Multiple Access Techniques for Intelligent and Multifunctional 6G: Tutorial, Survey, and Outlook
B. Clerckx;Yijie MaoZhaohui YangMingzhe ChenA. AlkhateebLiang LiuMin QiuJinhong YuanV. WongJ. Montojo
Proceedings of the IEEE Published 2024/01/02

Summary:

This article starts with an overview of orthogonal, physical-layer multicasting, space domain, power domain (PD), rate-splitting, code-domain MAs, MAs in other domains, and random access (RA), and highlights the importance of conducting research in universal MA (UMA) to shrink instead of grow the knowledge tree of MA schemes by providing a unified understanding of MA schemes across all resource dimensions.

A Closer Look at AUROC and AUPRC under Class Imbalance
Matthew B. A. McDermott;Lasse Hyldig HansenHaoran ZhangG. AngelottiJack Gallifant
ArXiv Published 2024/01/11

Summary:

This paper theoretically characterize the behavior of AUROC and AUPRC in the presence of model mistakes, establishing clearly that AUPRC is not generally superior in cases of class imbalance and shows that AUPRC can be a harmful metric.

Minimum Data Rate Maximization for Uplink Pinching-Antenna Systems
Sotiris A. Tegos;Panagiotis D. DiamantoulakisZhiguo DingG. Karagiannidis
IEEE Wireless Communications Letters Published 2024/12/18

Summary:

This letter addresses, for the first time, the uplink performance optimization of multi-user pinching-antenna (PA) systems, recently developed for next-generation wireless networks, and proposes an effective approach that separately optimizes the positions of the PAs and the resource allocation.

A novel efficient Rank-Revealing QR matrix and Schur decomposition method for big data mining and clustering (RRQR-SDM)
D. Paulraj;K. A. M. JunaidT. SethukarasiM. PremS. NeelakandanAdi AlhudhaifNorah M. Alnaim
Inf. Sci. Published 2024/02/01
Skillful joint probabilistic weather forecasting from marginals
Ferran Alet;Ilan PriceA. El-KadiDominic MastersStratis MarkouTom R. AnderssonJacklynn StottRemi LamMatthew WillsonAlvaro Sanchez-GonzalezPeter W. Battaglia
ArXiv Published 2025/06/12

Summary:

FGN is presented, a simple, scalable and flexible modeling approach which significantly outperforms the current state-of-the-art models and produces state-of-the-art ensemble forecasts as measured by a range of deterministic and probabilistic metrics.

Current practices and future direction of artificial intelligence in mathematics education: A systematic review
L. A. Awang;F. YusopM. Danaee
International Electronic Journal of Mathematics Education Published 2025/04/01

Summary:

This study conducts a systematic literature review (SLR) to investigate the applications and trends of AI in mathematics education by examining articles published in reputable journals indexed in Web of Science and Scopus.

Observer design method for nonlinear generalized systems with nonlinear algebraic constraints with applications
Shengya Meng;F. MengFan ZhangQi LiYu ZhangA. Zemouche
Autom. Published 2024/04/01
A Review of Graph Neural Networks in Epidemic Modeling
Zewen Liu;Guancheng WanB. A. PrakashMax S. Y. LauWei Jin
Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining Published 2024/03/28

Summary:

A comprehensive review of GNNs in epidemic tasks and methodologies is furnished, and hierarchical taxonomies for both epidemic tasks and methodologies are introduced, offering a trajectory of development within this domain.

When Hypergraph Meets Heterophily: New Benchmark Datasets and Baseline
Ming Li;Yongchun GuYi A WangYujie FangLu BaiXiaosheng ZhuangPietro Lio
Published 2025/04/11
Visualizing distributions of covariance matrices
T. Tokuda;Ben GoodrichI. MechelenA. GelmanF. Tuerlinckx
J. Data Sci. Stat. Vis. Published 2025/06/10

Summary:

This work demonstrates the approach for graphing distributions of covariance matrices on several models, including the Wishart, inverse-Wishart, and scaled inverse-Wishart families in different dimensions using a tableau of low-dimensional displays.

TabPFN-2.5: Advancing the State of the Art in Tabular Foundation Models
L'eo Grinsztajn;Klemens FlogeOscar KeyFelix BirkelPhilipp JundBrendan RoofBenjamin JagerDominik SafaricSimone AlessiA. HaylerMihir ManiumRose YuF. JablonskiShi Bin HooAnurag GargJake RobertsonMagnus BuhlerVladyslav MoroshanLennart PuruckerClara CornuLilly Charlotte WehrhahnAlessandro BonettoBernhard ScholkopfSauraj GambhirNoah HollmannFrank Hutter
ArXiv Published 2025/11/11

Summary:

For production use cases, a new distillation engine is introduced that converts TabPFN-2.5 into a compact MLP or tree ensemble, preserving most of its accuracy while delivering orders-of-magnitude lower latency and plug-and-play deployment.

Novel approximations to the fourth-order fractional Cahn–Hillard equations: Application to the Tantawy Technique and other two techniques with Yang transform
S. El-Tantawy;A. Al‐JohaniA. AlmuqrinAdnan KhanL. El-sherif
Journal of Low Frequency Noise, Vibration and Active Control Published 2025/03/19
fKAN: Fractional Kolmogorov-Arnold Networks with trainable Jacobi basis functions
A. Aghaei
ArXiv Published 2024/06/11

Summary:

The Fractional Kolmogorov-Arnold Network (fKAN), a novel neural network architecture that incorporates the distinctive attributes of KANs with a trainable adaptive fractional-orthogonal Jacobi function as its basis function, is presented.

Probabilistic surrogate modeling by Gaussian process: A review on recent insights in estimation and validation
A. Marrel;Bertrand Iooss
Reliab. Eng. Syst. Saf. Published 2024/03/01

Summary:

This paper focuses on the estimation of the Gaussian process covariance parameters by reviewing recent works on the analysis of the advantages and disadvantages of usual estimation methods, the most relevant validation criteria (for detecting poor estimation) and recent robust and corrective methods.

A Comprehensive Survey on Graph Reduction: Sparsification, Coarsening, and Condensation
Mohammad Hashemi;Shengbo GongJuntong NiWenqi FanB. A. PrakashWei Jin
Published 2024/01/29

Summary:

This survey aims to provide a comprehensive understanding of graph reduction methods, including graph sparsification, graph coarsening, and graph condensation, by establishing a unified definition for these methods and introducing a hierarchical taxonomy to categorize the challenges they address.

Deep Learning is Not So Mysterious or Different
A. G. Wilson
ArXiv Published 2025/03/03

Summary:

This work presents soft inductive biases as a key unifying principle in explaining these phenomena: rather than restricting the hypothesis space to avoid overfitting, embrace a flexible hypothesis space, with a soft preference for simpler solutions that are consistent with the data.

Exponential State Estimation for Delayed Competitive Neural Network Via Stochastic Sampled-Data Control with Markov Jump Parameters Under Actuator Failure
Yang Cao;Subhashri Alwarappan RavisankarA. ChandrasekarT. RadhikaKrzysztof Przybyszewski
Journal of Artificial Intelligence and Soft Computing Research Published 2024/07/01

Summary:

To ensure the exponentially mean-square stability of the delayed neural networks, the article constructs a Lyapunov-Krasovskii functional (LKF) that includes information about the bounds of the delay that is derived in the form of linear matrix inequalities by employing modified free matrix-based integral inequalities.

Memorization and Regularization in Generative Diffusion Models
Ricardo Baptista;Agnimitra DasguptaNikola B. KovachkiAssad A. OberaiAndrew M. Stuart
ArXiv Published 2025/01/27

Summary:

An analysis of the dynamical mechanism underlying memorization is presented, highlighting the need for regularization to avoid reproducing the analytically tractable minimizer; and laying the foundations for a principled understanding of how to regularize.

Combining physics-based and data-driven models: advancing the frontiers of research with Scientific Machine Learning
A. Quarteroni;P. GervasioFrancesco Regazzoni
ArXiv Published 2025/01/30

Summary:

The successful application of SciML to the simulation of the human cardiac function, a field of significant socioeconomic importance that poses numerous challenges on both the mathematical and computational fronts.

Development and Applications of a New Hybrid Weibull-Inverse Weibull Distribution
N. A. Noori;K. AbdullahM. Khaleel
Modern Journal of Statistics Published 2025/07/12
FourCastNet 3: A geometric approach to probabilistic machine-learning weather forecasting at scale
B. Bonev;Thorsten KurthA. MaheshM. BissonJean KossaifiK. KashinathAnima AnandkumarWilliam D. CollinsMichael S. PritchardAlexander Keller
ArXiv Published 2025/07/16

Summary:

Its computational efficiency, medium-range probabilistic skill, spectral fidelity, and rollout stability at subseasonal timescales make it a strong candidate for improving meteorological forecasting and early warning systems through large ensemble predictions.

Computational assessment of MgXH3 (X = Al, Sc and Zr) hydrides materials for hydrogen storage applications
S. Bahhar;A. TahiriA. JabarM. LouzazniM. IdiriH. Bioud
International Journal of Hydrogen Energy Published 2024/03/01
Production-inventory games: A new class of totally balanced combinatorial optimization games
L. A. Guardiola;A. MecaJ. Puerto
Games Econ. Behav. Published 2024/02/06

Summary:

This paper introduces a new class of cooperative games that arise from production-inventory problems, where several agents have to cover their demand over a finite time horizon and shortages are allowed.

Evaluation of University Professors Using the Spherical Fuzzy AHP and Grey MARCOS Multi-Criteria Decision-Making Model: A Case Study
Marko Radovanović;Stefan JovčićA. PetrovskiElif Cirkin
Spectrum of Decision Making and Applications Published 2025/01/23
Evolutionary dynamics of any multiplayer game on regular graphs
Chaoqian Wang;M. PercA. Szolnoki
Nature Communications Published 2024/01/22

Summary:

This research provides a mathematical framework to analyze multiplayer games with an arbitrary number of strategies on regular graphs by drawing an analogy with the Balls-and-Boxes problem, based on which it is shown that the local configuration of multiplayer games on graphs is equivalent to distributing k identical co-players among n distinct strategies.

Recent advances in Multi-objective Cuckoo Search Algorithm, its variants and applications
S. Makhadmeh;Mohammed A. AwadallahSofian KassaymehM. Al-BetarYousef K. SanjalaweShaimaa KoukaAnessa Al-Redhaei
Archives of Computational Methods in Engineering Published 2025/02/09

Summary:

This paper presents a comprehensive survey of 123 distinct variants of MOCSAs published in scientific journals and provides future research directions for MOCSA.

Who owns this site?

Article Galaxy Pages is a free service from Research Solutions, a company that offers access to content in collaboration with publishing partners, online repositories and discovery services.