October 2024 arXiv papers — page 71

Sebastian Töpfer, Sergio Tovar, Josué R. León Torres, Daniel Derr

Utilizing nonlinear interferometers for sensing with undetected light enables new sensing and imaging techniques in spectral ranges that are difficult to detect. To enhance this method for future applications, it is advantageous to extract both amplitude and phase information of an object. This study introduces two approaches for synthetic quantum holography

Yizhou Chi, Yizhang Lin, Sirui Hong, Duyi Pan

Automated Machine Learning (AutoML) approaches encompass traditional methods that optimize fixed pipelines for model selection and ensembling, as well as newer LLM-based frameworks that autonomously build pipelines. While LLM-based agents have shown promise in automating machine learning tasks, they often generate low-diversity and suboptimal code, even afte

Alexander F. Ritter, Filip Živanović

We prove that Floer theory induces a filtration by ideals on equivariant quantum cohomology of symplectic manifolds equipped with a $\mathbb{C}^*$-action. In particular, this gives rise to Hilbert-Poincar\'e polynomials on ordinary cohomology that depend on Floer theory. En route, the paper develops structural properties of filtrations on three versions of e

Hongru Cai, Yongqi Li, Wenjie Wang, Fengbin Zhu

Web agents have emerged as a promising direction to automate Web task completion based on user instructions, significantly enhancing user experience. Recently, Web agents have evolved from traditional agents to Large Language Models (LLMs)-based Web agents. Despite their success, existing LLM-based Web agents overlook the importance of personalized data (e.g

Robin Y. Park, Rhydian Windsor, Amir Jamaludin, Andrew Zisserman

We propose a general pipeline to automate the extraction of labels from radiology reports using large language models, which we validate on spinal MRI reports. The efficacy of our labelling method is measured on five distinct conditions: spinal cancer, stenosis, spondylolisthesis, cauda equina compression and herniation. Using open-source models, our method

Benedict Aaron Tjandra, Muhammed Razzak, Jannik Kossen, Kunal Handa

Large Language Models (LLMs) are known to hallucinate, whereby they generate plausible but inaccurate text. This phenomenon poses significant risks in critical applications, such as medicine or law, necessitating robust hallucination mitigation strategies. While recent works have proposed fine-tuning methods to teach LLMs to abstain from answering questions

Chao Yu, Qixin Tan, Hong Lu, Jiaxuan Gao

Preference-based reinforcement learning is an effective way to handle tasks where rewards are hard to specify but can be exceedingly inefficient as preference learning is often tabula rasa. We demonstrate that Large Language Models (LLMs) have native preference-learning capabilities that allow them to achieve sample-efficient preference learning, addressing

CMS Collaboration

A search for high-mass resonances decaying into a gluon, g, and two W bosons is presented. A Kaluza--Klein gluon, g$_\mathrm{KK}$, decaying in cascade via a scalar radion R, g$_\mathrm{KK}$ $\to$ gR $\to$ gWW, is considered. The final state studied consists of three large-radius jets, two of which contain the products of hadronically decaying W bosons, and t

Luca Boccioli, Mariam Gogilashvili, Jeremiah Murphy, Evan P. O'Connor

The explosion mechanism of a core-collapse supernova is a complex interplay between neutrino heating and cooling (including the effects of neutrino-driven convection), the gravitational potential, and the ram pressure of the infalling material. To analyze the post-bounce phase of a supernova, one can use the generalized Force Explosion Condition (FEC+), whic

Mads Bjerge Christensen

We study generating series encoding linking numbers between geodesics in arithmetic hyperbolic $3$-folds. We show that the series converge to functions on genus $2$ Siegel space and that certain explicit modifications have the transformation properties of genus $2$ Siegel modular forms of weight $2$. This is done by carefully analyzing the integral of the Ku

Thanasis Pittas, Ankit Pensia

Algorithmic robust statistics has traditionally focused on the contamination model where a small fraction of the samples are arbitrarily corrupted. We consider a recent contamination model that combines two kinds of corruptions: (i) small fraction of arbitrary outliers, as in classical robust statistics, and (ii) local perturbations, where samples may underg

Timothy Parker, Umberto Grandi, Emiliano Lorini

Responsibility is a key notion in multi-agent systems and in creating safe, reliable and ethical AI. However, most previous work on responsibility has only considered responsibility for single outcomes. In this paper we present a model for responsibility attribution in a multi-agent, multi-value setting. We also expand our model to cover responsibility antic

Victor Dubach

We study classical pattern counts in Mallows random permutations with parameters $(n,q_n)$, as $n\to\infty$. We focus on three different regimes for the parameter $q = q_n$. When $n^{3/2}(1-q)\to0$, we use coupling techniques to prove that pattern counts in Mallows random permutations satisfy a central limit theorem with the same asymptotic mean and variance

Haoqian Pan, Changhong Lu

In the wake of quantum computing advancements and quantum algorithmic progress, quantum algorithms are increasingly being employed to address a myriad of combinatorial optimization problems. Among these, the Independent Domination Problem (IDP), a derivative of the Domination Problem, has practical implications in various real-world scenarios. Despite this,

Letong Wang, Guy Blelloch, Yan Gu, Yihan Sun

Breadth-first Search (BFS) is one of the most important graph processing subroutines, especially for computing the unweighted distance. Many applications may require running BFS from multiple sources. Sequentially, when running BFS on a cluster of nearby vertices, a known optimization is using bit-parallelism. Given a subset of vertices with size $k$ and the

Azmine Toushik Wasi, Wahid Faisal, Taj Ahmad, Abdur Rahman

Climate change poses critical challenges globally, disproportionately affecting low-income countries that often lack resources and linguistic representation on the international stage. Despite Bangladesh's status as one of the most vulnerable nations to climate impacts, research gaps persist in Bengali-language studies related to climate change and NLP. To a

Nikita Nikolaev

We prove that formal WKB solutions of Schr\"odinger equations on Riemann surfaces are resurgent. Specifically, they are Borel summable in almost all directions and their Borel transforms admit endless analytic continuation away from a discrete subset of singularities. Our approach is purely geometric, relying on understanding the global geometry of complex f

Markus Müller, Ruslan Mushkaev

The surprisingly long-lasting oscillations observed in the dynamics of highly excited states of chains of Rydberg atoms defy the expectation that interacting systems should thermalize fast. The phenomenon is reminiscent of wavepackets in quantum billiards that trace classical periodic orbits. While analogs of the associated scarred eigenfunctions have been f

Tsachi Blau, Moshe Kimhi, Yonatan Belinkov, Alexander Bronstein

Fine-tuning Large Language Models (LLMs) typically involves updating at least a few billions of parameters. A more parameter-efficient approach is Prompt Tuning (PT), which updates only a few learnable tokens, and differently, In-Context Learning (ICL) adapts the model to a new task by simply including examples in the input without any training. When applyin

Zhaolin Ren, Runyu Zhang, Bo Dai, Na Li

Network Markov Decision Processes (MDPs), a popular model for multi-agent control, pose a significant challenge to efficient learning due to the exponential growth of the global state-action space with the number of agents. In this work, utilizing the exponential decay property of network dynamics, we first derive scalable spectral local representations for

David Ohlin, Anders Rantzer, Emma Tegling

This work considers infinite-horizon optimal control of positive linear systems applied to the case of network routing problems. We demonstrate the equivalence between Stochastic Shortest Path (SSP) problems and optimal control of a certain class of linear systems. This is used to construct a heuristic search framework for linear {positive} systems inspired

Y. Wolfovich, O. V. Gendelman

In addition to a common synchronization and/or localization behavior, a system of linearly coupled identical bistable Van der Pol (BVdP) oscillators can exhibit a "non-conventional" or "modal" synchronization. In two-DOF case, one can observe stable beatings attractor with synchronized amplitudes of the symmetric and antisymmetric modes. Current study demons

Mete Ismayilzada, Debjit Paul, Antoine Bosselut, Lonneke van der Plas

Creativity is the ability to produce novel, useful, and surprising ideas, and has been widely studied as a crucial aspect of human cognition. Machine creativity on the other hand has been a long-standing challenge. With the rise of advanced generative AI, there has been renewed interest and debate regarding AI's creative capabilities. Therefore, it is impera

Luccas Campos, Felipe Linares, Thyago S. R. Santos

We consider the $k$-dispersion generalized Benjamin-Ono ($k$-DGBO) equations. For nonlinearities with power $k \geq 4$, we establish local and global well-posedness results for the associated initial value problem (IVP) in both the critical and subcritical regimes, addressing sharp regularity in homogeneous and inhomogeneous Sobolev spaces. Additionally, our

Ali Baheri

The multi-armed bandit (MAB) problem is a foundational framework in sequential decision-making under uncertainty, extensively studied for its applications in areas such as clinical trials, online advertising, and resource allocation. Traditional MAB formulations, however, do not adequately capture scenarios where decisions are structured hierarchically, invo

Yuxian Gu, Hao Zhou, Fandong Meng, Jie Zhou

Knowledge distillation (KD) is widely used to train small, high-performing student language models (LMs) using large teacher LMs. While effective in fine-tuning, KD during pre-training faces efficiency, flexibility, and effectiveness issues. Existing methods either incur high computational costs due to online teacher inference, require tokenization matching

Louis Schatzki

In this short note we show that the ensemble $\{O \vert 0\rangle \langle 0 \vert O^\top \ \vert \ O \in \mathbb{O(d)}\}$, where $O$ is drawn from the Haar measure on $\mathbb{O}(d)$ cannot be distinguished from $t$ copies of a Haar random state unless $t = \Omega(\sqrt{d})$. Our proof has the benefit of exactly computing the trace distance, which scales as $

Zimeng Lyu, Amulya Saxena, Rohaan Nadeem, Hao Zhang

Stock return forecasting is a major component of numerous finance applications. Predicted stock returns can be incorporated into portfolio trading algorithms to make informed buy or sell decisions which can optimize returns. In such portfolio trading applications, the predictive performance of a time series forecasting model is crucial. In this work, we prop

Thomas Alazard, Chengyang Shao

Reducibility methods, aiming to simplify systems by conjugating them to those with constant coefficients, are crucial for studying the existence of quasiperiodic solutions. In KAM theory for PDEs, these methods help address the invertibility of linearized operators that arise in a Nash-Moser/KAM type scheme. The goal of this paper is to prove paradifferentia

Azmine Toushik Wasi, Wahid Faisal, Mst Rafia Islam, Mahathir Mohammad Bappy

Purpose: Bangladesh's legal system struggles with major challenges like delays, complexity, high costs, and millions of unresolved cases, which deter many from pursuing legal action due to lack of knowledge or financial constraints. This research seeks to develop a specialized Large Language Model (LLM) to assist in the Bangladeshi legal system. Methods: We

Hongyao Zhang, Bohang Sun

This thesis develops a Transformer model based on Whisper, which extracts melodies and chords from music audio and records them into ABC notation. A comprehensive data processing workflow is customized for ABC notation, including data cleansing, formatting, and conversion, and a mutation mechanism is implemented to increase the diversity and quality of train

Hamid Rahkooy

In this paper, we present a new algorithm for computing the linear recurrence relations of multi-dimensional sequences. Existing algorithms for computing these relations arise in computational algebra and include constructing structured matrices and computing their kernels. The challenging problem is to reduce the size of the corresponding matrices. In this

Zhiyi Pan, Guoqing Liu, Wei Gao, Thomas H. Li

The acquisition of inductive bias through point-level contrastive learning holds paramount significance in point cloud pre-training. However, the square growth in computational requirements with the scale of the point cloud poses a substantial impediment to the practical deployment and execution. To address this challenge, this paper proposes an Effective Po

Ujjval Bansal, Amit Sharma, Barbara Putz, Christoph Kirchlechner

Four-dimensional scanning transmission electron microscopy (4D-STEM) is a powerful tool that allows for the simultaneous acquisition of spatial and diffraction information, driven by recent advancements in direct electron detector technology. Although 4D-STEM has been predominantly developed for and used in conventional TEM and STEM, efforts are being made t

Sumith Reddy Anugu, Guodong Pang

We study the optimal scheduling problem for a Markovian multiclass queueing network with abandonment in the Halfin--Whitt regime, under the long run average (ergodic) risk sensitive cost criterion. The objective is to prove asymptotic optimality for the optimal control arising from the corresponding ergodic risk sensitive control (ERSC) problem for the limit

Anthony Chang, Narendra Singh, Marco Panesi

The most rigorous physical description of non-equilibrium gas dynamics is rooted in the numerical solution of the Boltzmann equation. Yet, the large number of degrees of freedom and the wide range of both spatial and temporal scales render these equations intractable for many relevant problems. This study constructs a reduced-order model for the Boltzmann eq

Tommaso Oss, Carlos E. Budde

Turing completeness has made Ethereum smart contracts attractive to blockchain developers and attackers alike. To increase code security, many tools can now spot most known vulnerabilities$-$at the cost of production efficiency. Recent studies show false-positive ratios over 99% in state-of-the-art technologies: this makes them impractical for use in industr

Juan Carlos Gonçalves-Dosantos, Laura Davila-Pena, Balbina Casas-Méndez

This paper presents a model for a vehicle routing problem in which customer demands are stochastic and vehicles are divided into compartments. The problem is motivated by the needs of certain agricultural cooperatives that produce various types of livestock food. The vehicles and their compartments have different capacities, and each compartment can only con

Anthony Degleris, Abbas El Gamal, Ram Rajagopal

We propose a GPU accelerated proximal message passing algorithm for solving contingency-constrained DC optimal power flow problems (OPF). We consider a highly general formulation of OPF that uses a sparse device-node model and supports a broad range of devices and constraints, e.g., energy storage and ramping limits. Our algorithm is a variant of the alterna

S. G. Baek, J. L. Terry, D. P. Stotler, B. Labombard

The ability to accurately model and predict neutral transport in the boundary plasma is important for tokamak operation. Nevertheless, validation of neutral transport models can be challenging due to the difficulty in measuring neutral particle distributions. Taking advantage of the localized neutral gas puff associated with the Gas Puff Imaging (GPI) diagno

Frederik L. Durhuus, Marco Beleggia, Cathrine Frandsen

For superparamagnetic nanoparticles (SMNPs), it is often claimed that the rapid thermal fluctuations of their magnetic moments negates the magnetic dipolar attraction, hence preventing aggregation in liquid suspension. However we find that this is a misconception. Using Langevin dynamics, we simulate SMNP pairs and the dimer clusters they form which is the s

Guodong Pang, Etienne Pardoux

We study the stochastic SIR epidemic model with infection-age dependent infectivity for which a measure-valued process is used to describe the ages of infection for each individual. We establish a functional law of large numbers (FLLN) and a functional central limit theorem (FCLT) for the properly scaled measure-valued processes together with the other epide

Aditya Nema, Sreejith Sreekumar, Mario Berta

We consider the problem of shared randomness-assisted multiple access channel (MAC) simulation for product inputs and characterize the one-shot communication cost region via almost-matching inner and outer bounds in terms of the smooth max-information of the channel, featuring auxiliary random variables of bounded size. The achievability relies on a rejectio

Paul Balister, Béla Bollobás, Marcelo Campos, Simon Griffiths

The $r$-colour Ramsey number $R_r(k)$ is the minimum $n \in \mathbb{N}$ such that every $r$-colouring of the edges of the complete graph $K_n$ on $n$ vertices contains a monochromatic copy of $K_k$. We prove, for each fixed $r \geqslant 2$, that $$R_r(k) \leqslant e^{-\delta k} r^{rk}$$ for some constant $\delta = \delta(r) > 0$ and all sufficiently large $k

Yiming Chen, Xianghu Yue, Chen Zhang, Xiaoxue Gao

Building on the success of large language models (LLMs), recent advancements such as GPT-4o have enabled real-time speech interactions through LLM-based voice assistants, offering a significantly improved user experience compared to traditional text-based interactions. However, the absence of benchmarks designed to evaluate these speech interaction capabilit

Chang Ma, Haiteng Zhao, Junlei Zhang, Junxian He

Large Language Models have demonstrated remarkable abilities in reasoning and planning by breaking down complex problems into sequential steps. Despite their success in various domains like mathematical problem-solving and coding, LLMs face challenges in ensuring reliable and optimal planning due to their inherent myopic nature of autoregressive decoding. Th

Kento Nishi, Rahul Ramesh, Maya Okawa, Mikail Khona

Knowledge Editing (KE) algorithms alter models' weights to perform targeted updates to incorrect, outdated, or otherwise unwanted factual associations. However, recent work has shown that applying KE can adversely affect models' broader factual recall accuracy and diminish their reasoning abilities. Although these studies give insights into the potential har

Kai Wang, Zekai Li, Zhi-Qi Cheng, Samir Khaki

Dataset distillation has demonstrated strong performance on simple datasets like CIFAR, MNIST, and TinyImageNet but struggles to achieve similar results in more complex scenarios. In this paper, we propose EDF (emphasizes the discriminative features), a dataset distillation method that enhances key discriminative regions in synthetic images using Grad-CAM ac

Yanfeng Hang, Cong Shen

In this work, we systematically study the differential systems governing loop-level wavefunction coefficients of conformally-coupled scalar field theory within a general power-law FRW cosmology. By utilizing the twisted cohomology, hyperplane arrangements, and IBP techniques, we derive the canonical differential equations for two-site one-loop bubble and tad

J. Elisenda Grigsby, Kathryn Lindsey

For any fixed feedforward ReLU neural network architecture, it is well-known that many different parameter settings can determine the same function. It is less well-known that the degree of this redundancy is inhomogeneous across parameter space. In this work, we discuss two locally applicable complexity measures for ReLU network classes and what we know abo

Ye Seol Lee, George Jackson, Amparo Galindo, Claire S. Adjiman

any practical multiobjective optimization (MOO) problems include discrete decision variables and/or nonlinear model equations and exhibit disconnected or smooth but nonconvex Pareto surfaces. Scalarization methods, such as the weighted-sum and sandwich (SD) algorithms, are common approaches to solving MOO problems but may fail on nonconvex or discontinuous P

R. Moradi, C. W. Wang, B. Zhang, Y. Wang

In this study, we present the pulse profile of the unique and the second brightest gamma-ray burst GRB 230307A, and analyze its temporal behavior using a joint GECAM--Fermi/GBM time-resolved spectral analysis. The utilization of GECAM data is advantageous as it successfully captured significant data during the pile-up period of the Fermi/GBM. We investigate

Ziv Scully, Laura Doval

We consider search problems with nonobligatory inspection and single-item or combinatorial selection. A decision maker is presented with a number of items, each of which contains an unknown price, and can pay an inspection cost to observe the item's price before selecting it. Under single-item selection, the decision maker must select one item; under combina

Samarth Kalluraya, Beichen Zhou, Yiannis Kantaros

In this paper, we consider teams of robots with heterogeneous skills (e.g., sensing and manipulation) tasked with collaborative missions described by Linear Temporal Logic (LTL) formulas. These LTL-encoded tasks require robots to apply their skills to specific regions and objects in a temporal and logical order. While existing temporal logic planning algorit

Yongjung Kim, Minjin Kim, Myungshin Im, Seo-Won Chang

We present a catalog of unobscured QSO candidates in the southern hemisphere from the early interim data of the KMTNet Synoptic Survey of Southern Sky (KS4). The KS4 data covers $\sim2500\,{\rm deg}^{2}$ sky area, reaching 5$\sigma$ detection limits of $\sim$22.1-22.7 AB mag in the $BVRI$ bands. Combining this with available infrared photometric data from th

Srujan Deolasee, Siva Kailas, Wenhao Luo, Katia Sycara

Informative path planning (IPP) is an important planning paradigm for various real-world robotic applications such as environment monitoring. IPP involves planning a path that can learn an accurate belief of the quantity of interest, while adhering to planning constraints. Traditional IPP methods typically require high computation time during execution, givi

Umang Mathur, David Mestel, Mahesh Viswanathan

The \emph{Entscheidungsproblem}, or the classical decision problem, asks whether a given formula of first-order logic is satisfiable. In this work, we consider an extension of this problem to regular first-order \emph{theories}, i.e., (infinite) regular sets of formulae. Building on the elegant classification of syntactic classes as decidable or undecidable

Kahlil Dozier, Justin Beltran, Kylie Berg, Hugo Matousek

Network verification (NWV), broadly defined as the verification of properties of distributed protocols used in network systems, cannot be efficiently solved on classical hardware via brute force. Prior work has developed a variety of methods that scale by observing a structure in the search space and then evaluating classes within the search space instead of

Thomas Lew, Marcus Greiff, Franck Djeumou, Makoto Suminaka

Model predictive control (MPC) algorithms can be sensitive to model mismatch when used in challenging nonlinear control tasks. In particular, the performance of MPC for vehicle control at the limits of handling suffers when the underlying model overestimates the vehicle's capabilities. In this work, we propose a risk-averse MPC framework that explicitly acco

David Andriot

We focus on minimally coupled (multi)field quintessence models, of thawing type, and their realistic solutions. In a model-independent manner, we describe analytically these cosmological solutions throughout the universe history. Starting with a kination - radiation domination phase, we obtain an upper bound on the scalar potential to guarantee an early kina

Shang-Jen Su, Shi-Yuan Wang, Matthieu R. Bloch, Zheshen Zhang

We derive a closed-form achievable rate for entanglement-assisted classical communication over a lossy thermal-noise bosonic channel, where the entanglement is in the form of a Two-Mode Squeezed Vacuum (TMSV) modulation restricted to Phase Shift Keying (PSK). The achievable rate is non-asymptotic in terms of the mean signal photon number, mean noise photon n

Robert Rosati, Tyson B. Littenberg

The Laser Interferometer Space Antenna (LISA) mission poses a difficult parameter estimation challenge: the sources will be so dense in both time and frequency that they all must be fit simultaneously in a `global fit'. Successful tests of global fit efforts on synthetic datasets have been recently reported, recovering extra-galactic black hole mergers and g

Vikrant Ashvinkumar, Aaron Bernstein, Adam Karczmarz

In the restricted shortest paths problem, we are given a graph $G$ whose edges are assigned two non-negative weights: lengths and delays, a source $s$, and a delay threshold $D$. The goal is to find, for each target $t$, the length of the shortest $(s,t)$-path whose total delay is at most $D$. While this problem is known to be NP-hard [Garey and Johnson, 197

X. Hernandez, Pavel Kroupa

Concerning recent published studies exploring the presence or otherwise of a gravitational anomaly at low accelerations in wide binary stars as observed by the {\it Gaia} satellite, the paper published by Cookson on the subject last year presents an interesting case. In that study, RMS values of binned relative internal velocities in 1D for wide binaries are

Yufan Zhang, Ronghai Wu, Michael Zaiser

The dynamics of dislocations can be formulated in terms of the evolution of continuous variables representing dislocation densities ('continuum dislocation dynamics'). We show for various variants of this approach that the resulting models can be envisaged in terms of the evolution of order-parameter like variables that strives to minimize a free energy func

M. Sh. Potashov, S. I. Blinnikov, E. I. Sorokina

The plasma opacity in stars depends mainly on the local state of matter (the density, temperature, and chemical composition at the point of interest), but in supernova ejecta it also depends on the expansion velocity gradient, because the Doppler effect shifts the spectral lines differently in different ejecta layers. This effect is known in the literature a

Nicholas Carlini, Milad Nasr

Scaling up language models has significantly increased their capabilities. But larger models are slower models, and so there is now an extensive body of work (e.g., speculative sampling or parallel decoding) that improves the (average case) efficiency of language model generation. But these techniques introduce data-dependent timing characteristics. We show

Prannay Kaul, Chengcheng Ma, Ismail Elezi, Jiankang Deng

We study two strange phenomena in auto-regressive Transformers: (1) the dominance of the first token in attention heads; (2) the occurrence of large outlier activations in the hidden states. We find that popular large language models, such as Llama attend maximally to the first token in 98% of attention heads, a behaviour we attribute to the softmax function

Yasha Ektefaie, Olivia Viessmann, Siddharth Narayanan, Drew Dresser

Protein inverse folding-that is, predicting an amino acid sequence that will fold into the desired 3D structure-is an important problem for structure-based protein design. Machine learning based methods for inverse folding typically use recovery of the original sequence as the optimization objective. However, inverse folding is a one-to-many problem where se

Md Meftahul Ferdaus, Mahdi Abdelguerfi, Elias Ioup, David Dobson

We introduce KANICE (Kolmogorov-Arnold Networks with Interactive Convolutional Elements), a novel neural architecture that combines Convolutional Neural Networks (CNNs) with Kolmogorov-Arnold Network (KAN) principles. KANICE integrates Interactive Convolutional Blocks (ICBs) and KAN linear layers into a CNN framework. This leverages KANs' universal approxima

J. Jorge, T. Barros, C. Premebida, M. Aleksandrov

Simultaneous Localization and Mapping (SLAM) is a key component of autonomous systems operating in environments that require a consistent map for reliable localization. SLAM has been a widely studied topic for decades with most of the solutions being camera or LiDAR based. Early LiDAR-based approaches primarily relied on 2D data, whereas more recent framewor

Miles Williams, George Chrysostomou, Nikolaos Aletras

Quantization and pruning are fundamental approaches for model compression, enabling efficient inference for language models. In a post-training setting, state-of-the-art quantization and pruning methods require calibration data, a small set of unlabeled examples. Conventionally, this is randomly sampled web text, aiming to reflect the model training data. Ho

Sagnick Mukherjee, Jonathan J. Fortney, Nicholas F. Wogan, David K. Sing

A primary goal of characterizing exoplanet atmospheres is to constrain planetary bulk properties, such as their metallicity, C/O ratio, and intrinsic heat. However, there are significant uncertainties in many aspects of atmospheric physics, such as the strength of vertical mixing. Here we use PICASO and the photochem model to explore how atmospheric chemistr

Ritesh Bachhar, Michael Pürrer, Stephen R. Green

As gravitational wave (GW) detector networks continue to improve in sensitivity, the demand on the accuracy of waveform models which predict the GW signals from compact binary coalescences is becoming more stringent. At high signal-to-noise ratios (SNRs) discrepancies between waveform models and the true solutions of Einstein's equations can introduce signif

Julius Rückin, David Morilla-Cabello, Cyrill Stachniss, Eduardo Montijano

Robots are frequently tasked to gather relevant sensor data in unknown terrains. A key challenge for classical path planning algorithms used for autonomous information gathering is adaptively replanning paths online as the terrain is explored given limited onboard compute resources. Recently, learning-based approaches emerged that train planning policies off

Jiaqi Hou

Let $\psi$ be an $L^2$-normalized Hecke-Maass form with a large spectral parameter $\lambda>0$ on a compact arithmetic congruence hyperbolic 3-manifold $X=\Gamma\backslash\mathrm{SL}(2,\mathbb{C})/\mathrm{SU}(2)$, and let $Y$ be a totally geodesic surface in $X$ with bounded diameter. The local $L^2$-bound for the restriction of $\psi$ to $Y$ is $\|\psi|_Y\|

J. W. Eatson, R. J. Parker, T. Lichtenberg

Recent research provides compelling evidence that the decay of short-lived radioisotopes (SLRs), such as 26Al, provided the bulk of energy for heating and desiccation of volatile-rich planetesimals in the early Solar System. However, it remains unclear whether the early Solar System was highly enriched relative to other planetary systems with similar formati

Camille Lagoin, Corentin Morin, Kirk Baldwin, Loren Pfeiffer

Enriching condensed-matter systems with quantum optical phenomena currently drives intense research efforts, particularly to introduce collective quantum correlations. Here we access this paradigm, by confining dipolar excitons in a nanoscopic lattice where long-range hopping, and nearest-neighbour dipolar repulsions, dress the Bose-Hubbard Hamiltonian. Long

İlker Işık, Ramazan Gokberk Cinbis, Ebru Aydin Gol

Language models lack the notion of interchangeable tokens: symbols that are semantically equivalent yet distinct, such as bound variables in formal logic. This limitation prevents generalization to larger vocabularies and hinders the model's ability to recognize alpha-equivalence, where renaming bound variables preserves meaning. We formalize this machine le

Zhuo Yao

Multi-Agent Path Finding (MAPF) has been widely studied in recent years. However, most existing MAPF algorithms assume that an agent occupies only a single grid in a grid-based map. This assumption limits their applicability in many real-world domains where agents have geometric shapes, rather than being point-like. Such agents, which can occupy multiple cel

Guoqi Yu, Yaoming Li, Xiaoyu Guo, Dayu Wang

Forecasting models are pivotal in a data-driven world with vast volumes of time series data that appear as a compound of vast Linear and Nonlinear patterns. Recent deep time series forecasting models struggle to utilize seasonal and trend decomposition to separate the entangled components. Such a strategy only explicitly extracts simple linear patterns like

Valentin Blomer, Jesse Thorner

We combine the relative trace formula with analytic methods to obtain zero density estimate for $L$-functions in various families of automorphic representations for $\mathrm{GL}(m)$. Applications include strong bounds for the average analytic rank of these $L$-functions at the central point and average equidistribution results for the imaginary parts of the

L. Bernal-González, M. C. Calderón-Moreno, J. López-Salazar, J. A. Prado-Bassas

In this paper, pointwise convergence, uniform convergence and compact convergence of sequences of holomorphic functions on an open subset of the complex plane are compared from a linear point of view. In fact, it is proved the existence of large linear algebras consisting, except for zero, of sequences of holomorphic functions tending to zero compactly but n

Roberto Robles, Chao Li, Sara Realista, Paulo Nuno Martinho

Scanning tunneling microscope data from a dinuclear Co(II) complex adsorbed on Au(111) are analysed using density functional theory calculations. We find that the interaction with the substrate substantially changes the geometry of the non-planar molecule. Its electronic states, however, remain fairly similar to those calculated for a gas-phase molecule. The

Suei-Wen Chen

This paper presents a method to establish functional inequalities via fuzzy decomposition on the state space, which generalizes earlier results dealing with exact partitions of the state space. Given a reversible Markov chain on a finite state space, we define its projection chain and restriction chains from classes of a fuzzy partition on the state space. T

Mary C. Tedeschi, Paola Ricaurte, Sridevi Ayloo, Joseph Corneli

At EuroPLoP 2024 Mary Tedeschi led the "AI Future Envisioning with PLACARD" focus group in Germany. Three conference attendees joined in the room while Sridevi, Paola, and Charles co-facilitated remotely via a web conference. The participants were introduced to a Futures Studies technique with the goal of capturing envisionments of Artificial Intelligence (A

Kieran Marray

Empirical researchers often estimate spillover effects by fitting linear or non-linear regression models to sampled network data. We show that common sampling schemes bias these estimates, potentially upwards, and derive biased-corrected estimators that researchers can construct from aggregate network statistics. Our results apply under different assumptions

Nathan Phelps, Daniel J. Lizotte, Douglas G. Woolford

When modelling data where the response is dichotomous and highly imbalanced, response-based sampling where a subset of the majority class is retained (i.e., undersampling) is often used to create more balanced training datasets prior to modelling. However, the models fit to this undersampled data, which we refer to as base models, generate predictions that a

Christopher D. Walker

This paper presents a Bayesian inference framework for a linear index threshold-crossing binary choice model that satisfies a median independence restriction. The key idea is that the model is observationally equivalent to a probit model with nonparametric heteroskedasticity. Consequently, Gibbs sampling techniques from Albert and Chib (1993) and Chib and Gr

Han Wang, Mukuntha Narayanan Sundararaman, Onur Gungor, Yu Xu

To improve relevance scoring on Pinterest Search, we integrate Large Language Models (LLMs) into our search relevance model, leveraging carefully designed text representations to predict the relevance of Pins effectively. Our approach uses search queries alongside content representations that include captions extracted from a generative visual language model

Kleber A. Oliveira, Henrique F. de Arruda, Yamir Moreno

We investigate how information-spreading mechanisms affect opinion dynamics and vice-versa via an agent-based simulation on adaptive social networks. First, we characterize the impact of reposting on user behavior with limited memory, a feature that introduces novel system states. Then, we build an experiment mimicking information-limiting environments seen

Kunal Pal, Kuntal Pal, Tapobrata Sarkar

Using the Israel-Darmois junction conditions, we obtain a class of regular dynamical interiors to the recently proposed black-bounce spacetimes which regularises the Schwarzschild singularity by introducing a regularisation parameter. We show that a regularised Friedmann-Lemaitre-Robertson-Walker like interior geometry can not be matched smoothly with the ex

Mahmoud Ali, Di Yang, François Brémond

Current vision-language foundation models, such as CLIP, have recently shown significant improvement in performance across various downstream tasks. However, whether such foundation models significantly improve more complex fine-grained action recognition tasks is still an open question. To answer this question and better find out the future research directi

Lilybelle Cowland Kellock

For $f(x)$ a separable polynomial of degree $d$ over a discretely valued field $K$, we describe how the cluster picture of $f(x)$ over $K$, in other words the set of tuples $\{(\mathrm{ord}(x_i-x_j),i,j) : 1\leq i< j \leq d \}$ where $x_1,\dots,x_d$ are the roots of $f(x)$, can be recovered without knowing the roots of $f(x)$ over $\bar{K}$. We construct an

Yunbum Kook, Matthew S. Zhang

We investigate the complexity of covariance matrix estimation for Gibbs distributions based on dependent samples from a Markov chain. We show that when $\pi$ satisfies a Poincar\'e inequality and the chain possesses a spectral gap, we can achieve similar sample complexity using MCMC as compared to an estimator constructed using i.i.d. samples, with potential

Ke Wang, Nikolaos Dimitriadis, Alessandro Favero, Guillermo Ortiz-Jimenez

Fine-tuning pre-trained models has become the standard approach to endow them with specialized knowledge, but it poses fundamental challenges. In particular, \textit{(i)} fine-tuning often leads to catastrophic forgetting, where improvements on a target domain degrade generalization on other tasks, and \textit{(ii)} merging fine-tuned checkpoints from dispar

Jirat Chiaranaipanich, Naiyarat Hanmatheekuna, Jitkapat Sawatphol, Krittamate Tiankanon

Large language models (LLMs) perform well on common tasks but struggle with generalization in low-resource and low-computation settings. We examine this limitation by testing various LLMs and specialized translation models on English-Thai machine translation and code-switching datasets. Our findings reveal that under more strict computational constraints, su

Junzhou Chen, Heqiang Huang, Ronghui Zhang, Nengchao Lyu

Ensuring safety in both autonomous driving and advanced driver-assistance systems (ADAS) depends critically on the efficient deployment of traffic sign recognition technology. While current methods show effectiveness, they often compromise between speed and accuracy. To address this issue, we present a novel real-time and efficient road sign detection networ

Sai Pushpak Nandanoori, Alok Kumar Bharati, Subhrajit Sinha, Soumya Kundu

With the emergence of low-inertia microgrids powered by inverter-based generation, there remains a concern about the operational resilience of these systems. Grid-forming inverters (GFMs), enabled by various device-level (primary) and system-level (secondary) control methods, are poised to play a significant role in achieving certain operational objectives,

M. V. Kornilov, V. S. Korolev, K. L. Malanchev, A. D. Lavrukhina

We present coniferest, an open source generic purpose active anomaly detection framework written in Python. The package design and implemented algorithms are described. Currently, static outlier detection analysis is supported via the Isolation forest algorithm. Moreover, Active Anomaly Discovery (AAD) and Pineforest algorithms are available to tackle active