May 2023 arXiv papers — page 34

Jesse Roberts

In this article we prove that the general transformer neural model undergirding modern large language models (LLMs) is Turing complete under reasonable assumptions. This is the first work to directly address the Turing completeness of the underlying technology employed in GPT-x as past work has focused on the more expressive, full auto-encoder transformer ar

Khushboo Dixit, S. Shajidul Haque, Soebur Razzaque

Quantum information theory has recently emerged as a flourishing area of research and quantum complexity, one of its powerful measures, is being applied for investigating complex systems in many areas of physics. Its application to practical physical situations, however, is still few and far between. Neutrino flavor oscillation is a widely studied physical p

Amir Jamaludin, Sarim Ather, Timor Kadir, Rhydian Windsor

This work introduces a simple deep-learning based method to delineate contours by `walking' along learnt unit vector fields. We demonstrate the effectiveness of our pipeline on the unique case of open contours on the task of delineating the sacroiliac joints (SIJs) in spinal MRIs. We show that: (i) 95% of the time the average root mean square error of the pr

Felix A. Wichmann, Robert Geirhos

Deep neural networks (DNNs) are machine learning algorithms that have revolutionised computer vision due to their remarkable successes in tasks like object classification and segmentation. The success of DNNs as computer vision algorithms has led to the suggestion that DNNs may also be good models of human visual perception. We here review evidence regarding

Saba Asaad, Hina Tabassum, Chongjun Ouyang, Ping Wang

Over-the-air federated learning (OTA-FL) is an emerging technique to reduce the computation and communication overload at the PS caused by the orthogonal transmissions of the model updates in conventional federated learning (FL). This reduction is achieved at the expense of introducing aggregation error that can be efficiently suppressed by means of receive

Dan Ley, Saumitra Mishra, Daniele Magazzeni

Counterfactual explanations have been widely studied in explainability, with a range of application dependent methods prominent in fairness, recourse and model understanding. The major shortcoming associated with these methods, however, is their inability to provide explanations beyond the local or instance-level. While many works touch upon the notion of a

Pietro Lesci, Yoshinari Fujinuma, Momchil Hardalov, Chao Shang

Sequence-to-sequence state-of-the-art systems for dialogue state tracking (DST) use the full dialogue history as input, represent the current state as a list with all the slots, and generate the entire state from scratch at each dialogue turn. This approach is inefficient, especially when the number of slots is large and the conversation is long. We propose

Siwei Wu, Xiangqing Shen, Rui Xia

The nodes in the commonsense knowledge graph (CSKG) are normally represented by free-form short text (e.g., word or phrase). Different nodes may represent the same concept. This leads to the problems of edge sparsity and node redundancy, which challenges CSKG representation and completion. On the one hand, edge sparsity limits the performance of graph repres

Charlie Street, Masoumeh Mansouri, Bruno Lacerda

Purpose of Review: To effectively synthesise and analyse multi-robot behaviour, we require formal task-level models which accurately capture multi-robot execution. In this paper, we review modelling formalisms for multi-robot systems under uncertainty, and discuss how they can be used for planning, reinforcement learning, model checking, and simulation. Rece

Georg Bökman, Fredrik Kahl

Many data symmetries can be described in terms of group equivariance and the most common way of encoding group equivariances in neural networks is by building linear layers that are group equivariant. In this work we investigate whether equivariance of a network implies that all layers are equivariant. On the theoretical side we find cases where equivariance

Nicolas Lanchier, Max Mercer

Allelopathy refers to a type $0/-$ biological interaction that is neutral for a so-called inhibitory species but detrimental for a so-called susceptible species. To model this type of interaction in a spatially-structured environment, Durrett and Levin introduced a variant of the multitype contact process in which the death rate of the susceptible species is

André Guerra, Eden Prywes

We investigate the following optimization problem: what is the least possible conformal capacity of a pair of linked curves in $S^3$? A natural conjecture, due to Gehring, Martin and Palka, is that the optimal value is attained by the standard Hopf link. We prove that this is the case under the assumption that each component of the link lies on a different s

István Porupsánszki

An edge-girth-regular graph $egr(n,k,g,\lambda)$ is a $k-$regular graph of order $n$, girth $g$ and with the property that each of its edges is contained in exactly $\lambda$ distinct $g-$cycles. We present new families of edge-girth regular graphs arising from generalized quadrangles and pencils of elliptic quadrics. An $egr(n, k, g, \lambda)$ is called ext

Sabit Hassan, Malihe Alikhani

Despite recent advancements, NLP models continue to be vulnerable to bias. This bias often originates from the uneven distribution of real-world data and can propagate through the annotation process. Escalated integration of these models in our lives calls for methods to mitigate bias without overbearing annotation costs. While active learning (AL) has shown

Pavel Gvozdevsky

We prove that an element from the Chevalley group of type $E_6$ or $E_7$ over a polynomial ring with coefficients in a small-dimensional ring can be reduced to an element of certain proper subsystem subgroup by a bounded number of elementary root elements. The bound is given explicitly. This result is an effective version of the early stabilisation of the co

Zhuoyan Luo, Yicheng Xiao, Yong Liu, Shuyan Li

This paper studies referring video object segmentation (RVOS) by boosting video-level visual-linguistic alignment. Recent approaches model the RVOS task as a sequence prediction problem and perform multi-modal interaction as well as segmentation for each frame separately. However, the lack of a global view of video content leads to difficulties in effectivel

Dinghuai Zhang, Hanjun Dai, Esmeralda S. Whitammer, Aaron Courville

Combinatorial optimization (CO) problems are often NP-hard and thus out of reach for exact algorithms, making them a tempting domain to apply machine learning methods. The highly structured constraints in these problems can hinder either optimization or sampling directly in the solution space. On the other hand, GFlowNets have recently emerged as a powerful

Sergio F. Yapur

This article examines a new approach to solving ordinary differential equations based on Fractional-Calculus theory. Poisson and Sturm-Liouville-type problems are studied, together with different boundary conditions. Each case is analyzed and compared concerning the Finite-Difference method outcome.

Caleb Ziems, Jane Dwivedi-Yu, Yi-Chia Wang, Alon Halevy

We present NormBank, a knowledge bank of 155k situational norms. This resource is designed to ground flexible normative reasoning for interactive, assistive, and collaborative AI systems. Unlike prior commonsense resources, NormBank grounds each inference within a multivalent sociocultural frame, which includes the setting (e.g., restaurant), the agents' con

Yuzhu Wang, Lechao Cheng, Manni Duan, Yongheng Wang

Knowledge distillation (KD) exploits a large well-trained model (i.e., teacher) to train a small student model on the same dataset for the same task. Treating teacher features as knowledge, prevailing methods of knowledge distillation train student by aligning its features with the teacher's, e.g., by minimizing the KL-divergence between their logits or L2 d

Yifan Du, Junyi Li, Tianyi Tang, Wayne Xin Zhao

In this paper, we propose a novel language model guided captioning approach, LAMOC, for knowledge-based visual question answering (VQA). Our approach employs the generated captions by a captioning model as the context of an answer prediction model, which is a Pre-trained Language model (PLM). As the major contribution, we leverage the guidance and feedback o

Kilian Pfeiffer, Ramin Khalili, Jörg Henkel

Federated learning (FL) is usually performed on resource-constrained edge devices, e.g., with limited memory for the computation. If the required memory to train a model exceeds this limit, the device will be excluded from the training. This can lead to a lower accuracy as valuable data and computation resources are excluded from training, also causing bias

M. Benitez Galan, L. Alvarez-Ruso, M. Rafi Alam, I. Ruiz Simo

In this work we study the production of $\Sigma$ and $\Lambda$ hyperons in strangeness changing $\Delta S = -1$ charged current interactions of muon antineutrinos on nuclear targets. At the nucleon level, besides quasielastic scattering we consider the inelastic mechanism in which a pion is produced alongside the hyperon. Its relevance for antineutrinos with

Armin Nurkanović, Jonathan Frey, Anton Pozharskiy, Moritz Diehl

The Finite Elements with Switch Detection (FESD) is a high-accuracy method for the numerical simulation and solution of optimal control problems subject to discontinuous ODEs. In this article, we extend the FESD method [Nurkanovi\'c et al., 2022] to the dynamic equations of multiple rigid bodies that exhibit state jumps due to impacts and Coulomb friction. T

Asahi Ushio, Fernando Alva-Manchego, Jose Camacho-Collados

Question and answer generation (QAG) consists of generating a set of question-answer pairs given a context (e.g. a paragraph). This task has a variety of applications, such as data augmentation for question answering (QA) models, information retrieval and education. In this paper, we establish baselines with three different QAG methodologies that leverage se

Michael Radica, Luis Welbanks, Néstor Espinoza, Jake Taylor

The future is now - after its long-awaited launch in December 2021, JWST began science operations in July 2022 and is already revolutionizing exoplanet astronomy. The Early Release Observations (ERO) program was designed to provide the first images and spectra from JWST, covering a multitude of science cases and using multiple modes of each on-board instrume

Matías Pizarro, Dorothea Kolossa, Asja Fischer

Adversarial attacks can mislead automatic speech recognition (ASR) systems into predicting an arbitrary target text, thus posing a clear security threat. To prevent such attacks, we propose DistriBlock, an efficient detection strategy applicable to any ASR system that predicts a probability distribution over output tokens in each time step. We measure a set

Jannik Kossen, Mark Collier, Basil Mustafa, Xiao Wang

We introduce Three Towers (3T), a flexible method to improve the contrastive learning of vision-language models by incorporating pretrained image classifiers. While contrastive models are usually trained from scratch, LiT (Zhai et al., 2022) has recently shown performance gains from using pretrained classifier embeddings. However, LiT directly replaces the i

Zhiyi Xue, Si Liu, Zhaodi Zhang, Yiting Wu

The robustness of deep neural networks (DNNs) is crucial to the hosting system's reliability and security. Formal verification has been demonstrated to be effective in providing provable robustness guarantees. To improve its scalability, over-approximating the non-linear activation functions in DNNs by linear constraints has been widely adopted, which transf

D. Gruner, S. A. Barnes, J. Weingrill

Gyrochronology allows the derivation of ages for cool main sequence stars from their observed rotation periods and masses, or a suitable proxy of the latter. It is increasingly well explored for FGK stars, but requires further measurements for older ages and K-M-type stars. Recent work has shown that the behavior of stellar spindown differs significantly fro

Yanyan Xu, Luis E. Olmos, David Mateo, Alberto Hernando

The urban spatial structure represents the distribution of public and private spaces in cities and how people move within them. While it usually evolves slowly, it can change fast during large-scale emergency events, as well as due to urban renewal in rapidly developing countries. This work presents an approach to delineate such urban dynamics in quasi-real-

Leonardo García-Heveling

We propose a new notion of singularity in General Relativity which complements the usual notions of geodesic incompleteness and curvature singularities. Concretely, we say that a spacetime has a volume singularity if there exist points whose future or past has arbitrarily small spacetime volume: in particular, smaller than a Planck volume. From a cosmologica

Jesper Amilon, Zafer Esen, Dilian Gurov, Christian Lidström

In deductive verification and software model checking, dealing with certain specification language constructs can be problematic when the back-end solver is not sufficiently powerful or lacks the required theories. One way to deal with this is to transform, for verification purposes, the program to an equivalent one not using the problematic constructs, and

ATLAS Collaboration

This Letter reports the observation of $WZ\gamma$ production and a measurement of its cross-section using 140.1 $\pm$ 1.2 fb$^{-1}$ of proton-proton collision data recorded at a center-of-mass energy of 13 TeV by the ATLAS detector at the Large Hadron Collider. The $WZ\gamma$ production cross-section, with both the $W$ and $Z$ bosons decaying leptonically, $

Srijoni Majumdar, Chuhao Qin, Evangelos Pournaras

Making Smart Cities more sustainable, resilient and democratic is emerging as an endeavor of satisfying hard constraints, for instance meeting net-zero targets. Decentralized multi-agent methods for socio-technical optimization of large-scale complex infrastructures such as energy and transport networks are scalable and more privacy-preserving by design. How

Philip Daniel Blocher, Karthik Chinni, Sivaprasad Omanakuttan, Pablo M. Poggi

The dynamical spreading of quantum information through a many-body system, typically called scrambling, is a complex process that has proven to be essential to describe many properties of out-of-equilibrium quantum systems. Scrambling can, in principle, be fully characterized via the use of out-of-time-ordered correlation functions, which are notoriously har

Gabriel Zapata, Tomás Urruzola, Oscar A. Sampayo, Lucía Duarte

The observation of neutrino oscillations and masses motivates the extension of the standard model with right handed neutrinos, leading to heavy neutrino states possibly in the electroweak scale, which could be impacted by new high-scale weakly coupled physics. A systematic tool for studying these interactions is the neutrino-extended standard model effective

Helena Häußler, Sebastian Schultheiß, Dirk Lewandowski

Search engines, such as Google, have a considerable impact on society. Therefore, undesirable consequences, such as retrieving incorrect search results, pose a risk to users. Although previous research has reported the adverse outcomes of web search, little is known about how search engine users evaluate those outcomes. In this study, we show which aspects o

Martin Holler, Erion Morina, Georg Schramm

This work is concerned with the identifiability of metabolic parameters from multi-region measurement data in quantitative PET imaging. It shows that, for the frequently used two-tissue compartment model and under reasonable assumptions, it is possible to uniquely identify metabolic tissue parameters from standard PET measurements, without the need of additi

Dennis Frauen, Valentyn Melnychuk, Stefan Feuerriegel

Causal inference from observational data is crucial for many disciplines such as medicine and economics. However, sharp bounds for causal effects under relaxations of the unconfoundedness assumption (causal sensitivity analysis) are subject to ongoing research. So far, works with sharp bounds are restricted to fairly simple settings (e.g., a single binary tr

Lindsay P. Walter, Joseph C. McKay, Bart Raeymaekers, Mathieu Francoeur

We model near-field thermal emission from metasurfaces structured as two-dimensional arrays of ellipsoidal SiC particles. The modeling approach is developed from fluctuational electrodynamics and is applicable to systems of ellipsoidal particles within the dipole limit. In all simulations, the radial lengths of particles are restricted to the range of 10 to

Alexandre Guével, Yue Meng, Christian Peco, Ruben Juanes

A Darcy-Cahn-Hilliard model coupled with damage is developed to describe multiphase-flow and fluid-driven fracturing in porous media. The model is motivated by recent experimental observations in Hele-Shaw cells of the fluid-driven fracturing of a synthetic porous medium with tunable fracture resistance. The model is derived from continuum thermodynamics and

Gengze Zhou, Yicong Hong, Qi Wu

Trained with an unprecedented scale of data, large language models (LLMs) like ChatGPT and GPT-4 exhibit the emergence of significant reasoning abilities from model scaling. Such a trend underscored the potential of training LLMs with unlimited language data, advancing the development of a universal embodied agent. In this work, we introduce the NavGPT, a pu

David Brandfonbrener, Ofir Nachum, Joan Bruna

In recent years, domains such as natural language processing and image recognition have popularized the paradigm of using large datasets to pretrain representations that can be effectively transferred to downstream tasks. In this work we evaluate how such a paradigm should be done in imitation learning, where both pretraining and finetuning data are trajecto

Philip Schär

We propose a new class of Markov chain Monte Carlo methods, called $k$-polar slice sampling ($k$-PSS), as a technical tool that interpolates between and extrapolates beyond uniform and polar slice sampling. By examining Wasserstein contraction rates and spectral gaps of $k$-PSS, we obtain strong quantitative results regarding its performance for different ki

Niccolò Tosato, Lorenzo Basile, Emanuele Ballarin, Giuseppe de Alteriis

The Backpropagation algorithm has often been criticised for its lack of biological realism. In an attempt to find a more biologically plausible alternative, the recently introduced Forward-Forward algorithm replaces the forward and backward passes of Backpropagation with two forward passes. In this work, we show that the internal representations obtained by

Zuzanna Kalinowska, Bartłomiej Dybiec

The Axelrod model is a cellular automaton which can be used to describe the emergence and development of cultural domains, where culture is represented by a fixed number of cultural features taking a discrete set of possible values (traits). The Axelrod model is based on two sociological phenomena: homophily (a tendency for individuals to form bonds with peo

Bei Li, Yi Jing, Xu Tan, Zhen Xing

Learning multiscale Transformer models has been evidenced as a viable approach to augmenting machine translation systems. Prior research has primarily focused on treating subwords as basic units in developing such systems. However, the incorporation of fine-grained character-level features into multiscale Transformer has not yet been explored. In this work,

Michael S. Chen, Yuezhi Mao, Andrew Snider, Prachi Gupta

Hydrogen bonding interactions with chromophores in chemical and biological environments play a key role in determining their electronic absorption and relaxation processes, which are manifested in their linear and multidimensional optical spectra. For chromophores in the condensed phase, the large number of atoms needed to simulate the environment has tradit

Peter R. Conwell, Kaushik Chakram, Valeria J. Villegas-Medina

In this paper we present a deterministic vertex spawning model that yields a scale-free network. The model specifies that a parent vertex produces a child vertex in a time interval approximately proportional to the current time and inversely proportional to the number of edges currently connected to the parent. Spawned offspring maintain an undirected edge w

Luc McCutcheon, Saber Fallah

Local-remote systems allow robots to execute complex tasks in hazardous environments such as space and nuclear power stations. However, establishing accurate positional mapping between local and remote devices can be difficult due to time delays that can compromise system performance and stability. Enhancing the synchronicity and stability of local-remote sy

Pratik Vadher, Giulia Sacco, Denys Nikolayev

This paper proposes a dual-band frequency scanning meandering microstrip leaky-wave antenna with linear polarization in the Ku-band and circular polarization in the K-band. This is achieved by making use of two spatial harmonics for radiation. The unit cell of the periodic microstrip antenna contains three meanders with mitred corners. To ensure circular pol

Fernando Argentieri, Bassam Fayad

We prove rotations-reducibility for close to constant quasi-periodic $SL(2,\mathbb{R})$ cocycles in one frequency in the finite regularity and smooth cases, and derive some applications to quasi-periodic Schr\"odinger operators.

Erick Lavoie

Conventional blockchains use consensus algorithms that totally order updates across all accounts, which is stronger than necessary to implement a replicated ledger. This makes updates slower and more expensive than necessary. More recent consensus-free replicated ledgers forego consensus algorithms, with significant increase in performance and decrease in in

Yves Annanias, Christofer Meinecke, Daniel Wiegreffe

The operation of open-cast lignite mines is a large intervention in nature, making the areas uninhabitable even after closing the mines without renaturation processes. Renaturation of these large areas requires a regional planning process which is tied to many conditions and restrictions, such as environmental protection laws. The related information is avai

Rahul Singh, Akshay Mete, Avik Kar, P. R. Kumar

Minimum variance controllers have been employed in a wide-range of industrial applications. A key challenge experienced by many adaptive controllers is their poor empirical performance in the initial stages of learning. In this paper, we address the problem of initializing them so that they provide acceptable transients, and also provide an accompanying fini

Merve Baksi, Hawoong Hong, Divine P. Kumah

We investigate the structural and transport properties of vacancy-ordered monoclinic superconducting $\mathrm{TiO}$ thin films grown by molecular beam epitaxy. The evolution of the crystal structure during growth is monitored by in-situ synchrotron X-ray diffraction. Long-range ordering of Ti and O vacancies in the disordered cubic phase stabilizes the vacan

Jan Ackermann, Christos Sakaridis, Fisher Yu

We present a simple and practical framework for anomaly segmentation called Maskomaly. It builds upon mask-based standard semantic segmentation networks by adding a simple inference-time post-processing step which leverages the raw mask outputs of such networks. Maskomaly does not require additional training and only adds a small computational overhead to in

Andrea Schioppa, Katja Filippova, Ivan Titov, Polina Zablotskaia

Influence functions (IF) have been seen as a technique for explaining model predictions through the lens of the training data. Their utility is assumed to be in identifying training examples "responsible" for a prediction so that, for example, correcting a prediction is possible by intervening on those examples (removing or editing them) and retraining the m

Kenzo Ishikawa

In this study, potential scatterings are formulated in experimental setups with Gaussian wave packets in accordance with a probability principle and associativity of products. A breaking of an associativity is observed in scalar products with stationary scattering states in a majority of short-range potentials. Due to the breaking, states of different energi

C. R. Hoffman, R. S. Lubna, E. Rubino, S. L. Tabor

Information on the $^{38}$S level scheme was expanded through experimental work utilizing a fusion-evaporation reaction and in-beam $\gamma$-ray spectroscopy. Prompt $\gamma$-ray transitions were detected by the Gamma-Ray Energy Tracking Array (GRETINA) and recoiling $^{38}$S residues were selected by the Fragment Mass Analayzer (FMA). Tools based on machine

Philippe Bernet, Joseph Chazalon, Edwin Carlinet, Alexandre Bourquelot

Linear objects convey substantial information about document structure, but are challenging to detect accurately because of degradation (curved, erased) or decoration (doubled, dashed). Many approaches can recover some vector representation, but only one closed-source technique introduced in 1994, based on Kalman filters (a particular case of Multiple Object

Kun Zhao, Bohao Yang, Chenghua Lin, Wenge Rong

The long-standing one-to-many issue of the open-domain dialogues poses significant challenges for automatic evaluation methods, i.e., there may be multiple suitable responses which differ in semantics for a given conversational context. To tackle this challenge, we propose a novel learning-based automatic evaluation metric (CMN), which can robustly evaluate

Marc Lafon, Elias Ramzi, Clément Rambour, Nicolas Thome

Out-of-distribution (OOD) detection is a critical requirement for the deployment of deep neural networks. This paper introduces the HEAT model, a new post-hoc OOD detection method estimating the density of in-distribution (ID) samples using hybrid energy-based models (EBM) in the feature space of a pre-trained backbone. HEAT complements prior density estimat

Gongye Liu, Haoze Sun, Jiayi Li, Fei Yin

Diffusion models have recently demonstrated an impressive ability to address inverse problems in an unsupervised manner. While existing methods primarily focus on modifying the posterior sampling process, the potential of the forward process remains largely unexplored. In this work, we propose Shortcut Sampling for Diffusion(SSD), a novel approach for solvin

Felix Hartmann, Stefano Scali, Janet Anders

Thermal equilibrium properties of nanoscale systems deviate from standard macroscopic predictions due to a non-negligible coupling to the environment. For anisotropic three-dimensional materials, we derive the mean force corrections to the equilibrium state of a classical spin vector. The result is valid at arbitrary coupling strength. Specifically, we consi

Yuchen Bai, Jean-Baptiste Durand, Grégoire Vincent, Florence Forbes

LiDAR (Light Detection and Ranging) has become an essential part of the remote sensing toolbox used for biosphere monitoring. In particular, LiDAR provides the opportunity to map forest leaf area with unprecedented accuracy, while leaf area has remained an important source of uncertainty affecting models of gas exchanges between the vegetation and the atmosp

Hao Guo, Wanxin Li, Mark Nejad

Blockchain-based IoT systems can manage IoT devices and achieve a high level of data integrity, security, and provenance. However, incorporating the existing consensus protocols in many IoT systems limits scalability and leads to high computational cost and network latency. We propose a hierar-chical and location-aware consensus protocol for IoI-blockchain a

Mahmut Bicer, Krishna C Balram

Guiding and manipulating GHz frequency acoustic waves in $\mu$m-scale waveguides and resonators opens up new degrees of freedom to manipulate radio frequency (RF) signals in chip-scale platforms. A critical requirement for enabling high-performance devices is the demonstration of low acoustic dissipation in these highly confined geometries. In this work, we

Ruibo Liu, Ruixin Yang, Chenyan Jia, Ge Zhang

Social alignment in AI systems aims to ensure that these models behave according to established societal values. However, unlike humans, who derive consensus on value judgments through social interaction, current language models (LMs) are trained to rigidly replicate their training corpus in isolation, leading to subpar generalization in unfamiliar scenarios

Enzo Monino, Pierre-François Loos

In recent years, Green's function methods have garnered considerable interest due to their ability to target both charged and neutral excitations. Among them, the well-established $GW$ approximation provides accurate ionization potentials and electron affinities and can be extended to neutral excitations using the Bethe-Salpeter equation (BSE) formalism. Her

Shiyue Zhang, Shijie Wu, Ozan Irsoy, Steven Lu

Autoregressive language models are trained by minimizing the cross-entropy of the model distribution Q relative to the data distribution P -- that is, minimizing the forward cross-entropy, which is equivalent to maximum likelihood estimation (MLE). We have observed that models trained in this way may "over-generalize", in the sense that they produce non-huma

Vineet Bhat, Preethi Jyothi, Pushpak Bhattacharyya

Conversational speech often consists of deviations from the speech plan, producing disfluent utterances that affect downstream NLP tasks. Removing these disfluencies is necessary to create fluent and coherent speech. This paper presents DisfluencyFixer, a tool that performs speech-to-speech disfluency correction in English and Hindi using a pipeline of Autom

Fabio Anselmi, Mauro Castelli, Alberto d'Onofrio, Luca Manzoni

Geometric Semantic Geometric Programming (GSGP) is one of the most prominent Genetic Programming (GP) variants, thanks to its solid theoretical background, the excellent performance achieved, and the execution time significantly smaller than standard syntax-based GP. In recent years, a new mutation operator, Geometric Semantic Mutation with Local Search (GSM

Apoorva Nagar, Shamik Gupta

We review recent work on systems with multiple interacting-particles having the dynamical feature of stochastic resetting. The interplay of time scales related to inter-particle interactions and resetting leads to a rich behavior, both static and dynamic. The presence of multiple particles also opens up a new possibility for the resetting dynamics itself, na

Matthias Lindemann, Alexander Koller, Ivan Titov

Seq2seq models have been shown to struggle with compositional generalization in semantic parsing, i.e. generalizing to unseen compositions of phenomena that the model handles correctly in isolation. We phrase semantic parsing as a two-step process: we first tag each input token with a multiset of output tokens. Then we arrange the tokens into an output seque

Josef Spjut, Arjun Madhusudan, Benjamin Watson, Seth Schneider

Gamers use a variety of different display sizes, though for PC gaming in particular, monitors in the 24 to 27 inch size range have become most popular. Particularly popular among many PC gamers, first person shooter (FPS) games represent a genre where hand-eye coordination is particularly central to the player's performance in game. In a carefully designed p

Nicolas Kaufmann, Yann Alibert

Context. The effects of planetesimal fragmentation on planet formation has been studied by various models on single embryos therefore neglecting concurrent effects mostly in the outer disk. They show that planetesimal fragmentation can both hinder or aid planet formation due to the introduction of competing effects, namely speeding up accretion and depleting

Amal M A Alghamdi, Nicolai A B Riis, Babak M Afkham, Felipe Uribe

Inverse problems, particularly those governed by Partial Differential Equations (PDEs), are prevalent in various scientific and engineering applications, and uncertainty quantification (UQ) of solutions to these problems is essential for informed decision-making. This second part of a two-paper series builds upon the foundation set by the first part, which i

Philipp Mohr, Syed Aizaz Ali Shah, Gerhard Bauch

An implementation-efficient finite alphabet decoder for polar codes relying on coarsely quantized messages and low-complexity operations is proposed. Typically, finite alphabet decoding performs concatenated compression operations on the received channel messages to aggregate compact reliability information for error correction. These compression operations

Nicolai A B Riis, Amal M A Alghamdi, Felipe Uribe, Silja L Christensen

This paper introduces CUQIpy, a versatile open-source Python package for computational uncertainty quantification (UQ) in inverse problems, presented as Part I of a two-part series. CUQIpy employs a Bayesian framework, integrating prior knowledge with observed data to produce posterior probability distributions that characterize the uncertainty in computed s

Hayeon Lee, Sohyun An, Minseon Kim, Sung Ju Hwang

Distillation-aware Neural Architecture Search (DaNAS) aims to search for an optimal student architecture that obtains the best performance and/or efficiency when distilling the knowledge from a given teacher model. Previous DaNAS methods have mostly tackled the search for the neural architecture for fixed datasets and the teacher, which are not generalized w

Matt Grenander, Shay B. Cohen, Mark Steedman

We propose a sentence-incremental neural coreference resolution system which incrementally builds clusters after marking mention boundaries in a shift-reduce method. The system is aimed at bridging two recent approaches at coreference resolution: (1) state-of-the-art non-incremental models that incur quadratic complexity in document length with high computat

Timo Bröhl, Klaus Lehnertz

Constructing networks from empirical time series data is often faced with the as yet unsolved issue of how to avoid potentially superfluous network constituents. Such constituents can result, e.g., from spatial and temporal oversampling of the system's dynamics, and neglecting them can lead to severe misinterpretations of network characteristics ranging from

Laurent Orseau, Marcus Hutter, Levi H. S. Lelis

Levin Tree Search (LTS) is a search algorithm that makes use of a policy (a probability distribution over actions) and comes with a theoretical guarantee on the number of expansions before reaching a goal node, depending on the quality of the policy. This guarantee can be used as a loss function, which we call the LTS loss, to optimize neural networks repres

Tianyi Tang, Yushuo Chen, Yifan Du, Junyi Li

People often imagine relevant scenes to aid in the writing process. In this work, we aim to utilize visual information for composition in the same manner as humans. We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration. First, we imagine the scene based on the text: we use

Leon Migus, Julien Salomon, Patrick Gallinari

Forecasting physical signals in long time range is among the most challenging tasks in Partial Differential Equations (PDEs) research. To circumvent limitations of traditional solvers, many different Deep Learning methods have been proposed. They are all based on auto-regressive methods and exhibit stability issues. Drawing inspiration from the stability pro

Sohyun An, Hayeon Lee, Jaehyeong Jo, Seanie Lee

Existing NAS methods suffer from either an excessive amount of time for repetitive sampling and training of many task-irrelevant architectures. To tackle such limitations of existing NAS methods, we propose a paradigm shift from NAS to a novel conditional Neural Architecture Generation (NAG) framework based on diffusion models, dubbed DiffusionNAG. Specifica

Karim Achouri, Mintae Chung, Andrei Kiselev, Olivier J. F. Martin

It has been observed that achiral nano-particles, such as flat helices, may be subjected to an optical torque even when illuminated by normally incident linearly polarized light. However, the origin of this fascinating phenomenon has so far remained mostly unexplained. We therefore propose an exhaustive discussion that provides a clear and rigorous explanati

Smitha Milli, Micah Carroll, Yike Wang, Sashrika Pandey

In a pre-registered algorithmic audit, we found that, relative to a reverse-chronological baseline, Twitter's engagement-based ranking algorithm amplifies emotionally charged, out-group hostile content that users say makes them feel worse about their political out-group. Furthermore, we find that users do \emph{not} prefer the political tweets selected by th

Bertrand Eynard, Elba Garcia-Failde, Paolo Gregori, Danilo Lewanski

Jackiw-Teitelboim dilaton-quantum-gravity localizes on a double-scaled random-matrix model, whose perturbative free energy is an asymptotic series. Understanding the resurgent properties of this asymptotic series, including its completion into a full transseries, requires understanding the nonperturbative instanton sectors of the matrix model for Jackiw-Teit

Kenzo Ishikawa

Orthogonality of eigenstates of different energies and its implications in potential scattering are unlabeled. Scalar products of scattering states of different energies are found to have finite non-orthogonal terms in potentials of finite widths. Their superpositions have time-dependent norms, and are not suitable for isolate states. In these systems, a per

Marius Mosbach, Tiago Pimentel, Shauli Ravfogel, Dietrich Klakow

Few-shot fine-tuning and in-context learning are two alternative strategies for task adaptation of pre-trained language models. Recently, in-context learning has gained popularity over fine-tuning due to its simplicity and improved out-of-domain generalization, and because extensive evidence shows that fine-tuned models pick up on spurious correlations. Unfo

Bum Chul Kwon, Nandana Mihindukulasooriya

Pre-trained transformer-based language models are becoming increasingly popular due to their exceptional performance on various benchmarks. However, concerns persist regarding the presence of hidden biases within these models, which can lead to discriminatory outcomes and reinforce harmful stereotypes. To address this issue, we propose Finspector, a human-ce

Jiwen Yu, Xuanyu Zhang, Youmin Xu, Jian Zhang

Current image steganography techniques are mainly focused on cover-based methods, which commonly have the risk of leaking secret images and poor robustness against degraded container images. Inspired by recent developments in diffusion models, we discovered that two properties of diffusion models, the ability to achieve translation between two images without

Lena Cabrera, Jan Niehues

Neural machine translation (NMT) models often suffer from gender biases that harm users and society at large. In this work, we explore how bridging the gap between languages for which parallel data is not available affects gender bias in multilingual NMT, specifically for zero-shot directions. We evaluate translation between grammatical gender languages whic

Yunqing Zhao, Tianyu Pang, Chao Du, Xiao Yang

Large vision-language models (VLMs) such as GPT-4 have achieved unprecedented performance in response generation, especially with visual inputs, enabling more creative and adaptable interaction than large language models such as ChatGPT. Nonetheless, multimodal generation exacerbates safety concerns, since adversaries may successfully evade the entire system

Christoph Koutschan, Bernhard Moser, Anton Ponomarchuk, Josef Schicho

A piecewise linear function can be described in different forms: as an arbitrarily nested expression of $\min$- and $\max$-functions, as a difference of two convex piecewise linear functions, or as a linear combination of maxima of affine-linear functions. In this paper, we provide two main results: first, we show that for every piecewise linear function the

Chen Chen, Chao-Han Huck Yang, Kai Li, Yuchen Hu

In this work, we introduce S4M, a new efficient speech separation framework based on neural state-space models (SSM). Motivated by linear time-invariant systems for sequence modeling, our SSM-based approach can efficiently model input signals into a format of linear ordinary differential equations (ODEs) for representation learning. To extend the SSM techniq

Marco Erba, Paolo Perinotti, Davide Rolino, Alessandro Tosini

The core of Heisenberg's heuristic argument for the uncertainty principle, involving the famous $\gamma$-ray microscope $\textit{Gedankenexperiment}$, hinges upon the existence of measurements that irreversibly alter the state of the system on which they are acting, causing an irreducible disturbance on subsequent measurements. The argument was put forward t