Search

Generative artificial intelligence (AI) technology is revolutionizing the computing industry. Not only its applications have broadened to various sectors but also poses new system design and optimization opportunities. The technology is capable of understanding and responding in multiple modalities. However, the advanced capability currently comes with significant system resource demands. To sustainably scale generative AI capabilities to billions of users in the world, inference must be fast and efficient. This paper pinpoints key system design and optimization opportunities by characterizing a family of emerging multi-modal generation models on real systems. Auto-regressive token generation is a critical latency performance bottleneck, typically dominated by GPU idle time. In addition to memory-intensive attention across the generative AI models, linear operations constitute significant inference latency due to the feed forward networks in Transformer-based models. We demonstrate that state-of-the-art optimization levers, spanning from applications to system software and hardware, set a 3.88x better baseline., Comment: 13 pages including references. 8 Figures. Under review to HPCA 2025 Industry Track

Lifelong imitation learning for manipulation tasks poses significant challenges due to distribution shifts that occur in incremental learning steps. Existing methods often focus on unsupervised skill discovery to construct an ever-growing skill library or distillation from multiple policies, which can lead to scalability issues as diverse manipulation tasks are continually introduced and may fail to ensure a consistent latent space throughout the learning process, leading to catastrophic forgetting of previously learned skills. In this paper, we introduce M2Distill, a multi-modal distillation-based method for lifelong imitation learning focusing on preserving consistent latent space across vision, language, and action distributions throughout the learning process. By regulating the shifts in latent representations across different modalities from previous to current steps, and reducing discrepancies in Gaussian Mixture Model (GMM) policies between consecutive learning steps, we ensure that the learned policy retains its ability to perform previously learned tasks while seamlessly integrating new skills. Extensive evaluations on the LIBERO lifelong imitation learning benchmark suites, including LIBERO-OBJECT, LIBERO-GOAL, and LIBERO-SPATIAL, demonstrate that our method consistently outperforms prior state-of-the-art methods across all evaluated metrics., Comment: Submitted to ICRA2025

Data privacy and authentication are two main security requirements for remote access and cloud services. While QKD has been explored to address data privacy concerns, oftentimes its use is separate from the client authentication protocol despite implicitly providing authentication. Here, we present a quantum authentication key expansion (QAKE) protocol that (1) integrates both authentication and key expansion within a single protocol, and (2) provides key recycling property -- allowing all authentication keys to be reused. We analyse the security of the protocol in a QAKE framework adapted from a classical authentication key exchange (AKE) framework, providing separate security conditions for authentication and data privacy. An experimental implementation of the protocol, with appropriate post-selection, was performed to demonstrate its feasibility., Comment: 71 pages, comments are welcome

In this work, fission of moderately excited compound nucleus $^{220}$Ra${^\star}$ produced in $^{12}$C+$^{208}$Pb reaction at E$_{\rm lab}$ = 81.9, 75.8, and 67.5 MeV has been studied. 28 fission fragments, within the mass range 60$\leq$A$\leq$141, have been identified based on their characteristic decay $\gamma$-lines and half-lives. The yields of different fission fragments have been analyzed to generate isotopic and isobaric yield distributions. The value of the mass dispersion parameter, $\sigma^2_A$, is found to be 2.93 and 2.65 for Antimony (Sb) isotope at excitation energy E$^{\star}$ = 45.4 and 39.6 MeV, and 1.24 for Indium (In) isotope at E$^{\star}$ = 45.4 MeV. The charge dispersion parameter $\sigma_z$ for Sb is estimated to be 0.769 and 0.714 at E$^{\star}$ = 45.4 and 39.6 MeV, respectively. For In isotopes, the value of $\sigma_z$ is estimated to be 0.430 at E$^{\star}$ = 45.4 MeV. The value of mass and charge dispersion parameters for Sb and In isotopes display good agreement with the values reported in the literature for similar systems. It has been found that $^{99m}$Tc and $^{111}$In, medically important isotopes, populate in this system. The present investigations suggest that fission is a dominating mode of deexcitation of the compound nucleus, even at entrance channel energies slightly above the Coulomb barrier.

The Hellinger distance between quantum states is a significant measure in quantum information theory, known for its Riemannian and monotonic properties. It is also easier to compute than the Bures distance, another measure that shares these properties. In this work, we derive the mean and variance of the Hellinger distance between pairs of density matrices, where one or both matrices are random. Along the way, we also obtain exact results for the mean affinity and mean square affinity. The first two cumulants of the Hellinger distance allow us to propose an approximation for the corresponding probability density function based on the gamma distribution. Our analytical results are corroborated through Monte Carlo simulations, showing excellent agreement., Comment: 8 pages, 5 figures

The recently discovered bright transient black hole candidate Swift J1727.8-1613 is studied in a broad energy range ($0.5-79$ keV) using combined NICER and NuSTAR data on 29 August 2023. A promonient type-C Quasi-Periodic Oscillation (QPO) at $0.89 \pm 0.01$ Hz with its harmonic was observed in NICER data of $0.5-10$ keV. Interestingly, the harmonic becomes weaker in the lower energy bands ($0.5-1$ & $1-3$ keV). We also report the first detection of a soft time-lag of $0.014 \pm 0.001$ s at the QPO frequency between harder ($3-10$ kev) and softer ($0.5-3$ keV) band photons observed with the NICER/XTI instrument. This indicates that the inclination of the accretion disk in the binary system might be high. From the detailed spectral analysis with the relxill reflection model, we found the disk inclination angle of source to be $\sim 85^\circ$. We discuss how the accretion flow configuration inferred from spectral analysis can help us to understand the origin of QPOs and soft lag in this source., Comment: 10 pages, 4 figures, 1 table (in press ApJ)

We present a time-domain inspiral-merger-ringdowm (IMR) waveform model ESIGMAHM constructed within a framework we named ESIGMA for coalescing binaries of spinning black holes on moderately eccentric orbits (Huerta et al. (2018) [Phys. Rev. D 97, 024031]). We now include the effect of black hole spins on the dynamics of eccentric binaries, as well as model sub-dominant waveform harmonics emitted by them. The inspiral evolution is described by a consistent combination of latest results from post-Newtonian theory, self-force, and black hole perturbation theory. We assume that these moderately eccentric binaries radiate away most of their orbital eccentricity before merger, and seamlessly connect the eccentric inspiral with a numerical relativity based surrogate waveform model for mergers of spinning binaries on quasi-circular orbits. We validate ESIGMAHM against eccentric Numerical Relativity simulations, and also against contemporary effective-one-body and phenomenological models in the quasi-circular limit. We find that ESIGMAHM achieves match values greater than $99\%$ for quasi-circular spin-aligned binaries with mass ratios up to $8$, and above $97\%$ for non-spinning and spinning eccentric systems with small or positively aligned spins. Using IMRESIGMA, we quantify the impact of orbital eccentricity on GW signals, showing that next-generation detectors can detect eccentric sources up to $10\%$ louder than quasi-circular ones. We also show that current templated LIGO-Virgo searches will lose more than $10\%$ of optimal SNR for about $20\%$ of all eccentric sources by using only quasi-circular waveform templates. The same will result in a $25\%$ loss in detection rate for eccentric sources with mass ratios $m_1/m_2\geq 4$. Our results highlight the need for including eccentricity and higher-order modes in GW source models and searches for asymmetric eccentric BBH signals.

A threshold dynamics model of grain growth that accounts for the anisotropy in the grain boundary energy has been used to simulate experimentally observed grain growth of polycrystalline Ni. The simulation reproduces several aspects of the observed microstructural evolution that are not found in the results of simulations assuming isotropic properties. For example, the relative areas of the lowest-energy twin boundaries increase as the grains grow and the average grain boundary energy decreases with grain growth. This decrease in energy occurs because the population of higher-energy grain boundaries decreases while the population of lower-energy boundaries increases as the total grain boundary area decreases. This phenomenon emerges from the assumption of anisotropic grain boundary energies without modification of the energy minimizing algorithm. These findings are consistent with the observation that, in addition to the decrease in grain boundary area, additional energy is dissipated during grain growth by a decrease in the average grain boundary energy.

In this thesis, we consider an $N$-dimensional Ornstein-Uhlenbeck (OU) process satisfying the linear stochastic differential equation $d\mathbf x(t) = - \mathbf B\mathbf x(t) dt + \boldsymbol \Sigma d \mathbf w(t).$ Here, $\mathbf B$ is a fixed $N \times N$ circulant friction matrix whose eigenvalues have positive real parts, $\boldsymbol \Sigma$ is a fixed $N \times M$ matrix. We consider a signal propagation model governed by this OU process. In this model, an underlying signal propagates throughout a network consisting of $N$ linked sensors located in space. We interpret the $n$-th component of the OU process as the measurement of the propagating effect made by the $n$-th sensor. The matrix $\mathbf B$ represents the sensor network structure: if $\mathbf B$ has first row $(b_1 \ , \ \dots \ , \ b_N),$ where $b_1>0$ and $b_2 \ , \ \dots \ ,\ b_N \le 0,$ then the magnitude of $b_p$ quantifies how receptive the $n$-th sensor is to activity within the $(n+p-1)$-th sensor. Finally, the $(m,n)$-th entry of the matrix $\mathbf D = \frac{\boldsymbol \Sigma \boldsymbol \Sigma^\text T}{2}$ is the covariance of the component noises injected into the $m$-th and $n$-th sensors. For different choices of $\mathbf B$ and $\boldsymbol \Sigma,$ we investigate whether Cyclicity Analysis enables us to recover the structure of network. Roughly speaking, Cyclicity Analysis studies the lead-lag dynamics pertaining to the components of a multivariate signal. We specifically consider an $N \times N$ skew-symmetric matrix $\mathbf Q,$ known as the lead matrix, in which the sign of its $(m,n)$-th entry captures the lead-lag relationship between the $m$-th and $n$-th component OU processes. We investigate whether the structure of the leading eigenvector of $\mathbf Q,$ the eigenvector corresponding to the largest eigenvalue of $\mathbf Q$ in modulus, reflects the network structure induced by $\mathbf B.$, Comment: Ph.D. thesis successfully defended and deposited in July 2024. To appear in the IDEALS repository

Large Language Models (LLMs) are widely used in healthcare, but limitations like hallucinations, incomplete information, and bias hinder their reliability. To address these, researchers released the Build Your Own expert Bot (BYOeB) platform, enabling developers to create LLM-powered chatbots with integrated expert verification. CataractBot, its first implementation, provides expert-verified responses to cataract surgery questions. A pilot evaluation showed its potential; however the study had a small sample size and was primarily qualitative. In this work, we conducted a large-scale 24-week deployment of CataractBot involving 318 patients and attendants who sent 1,992 messages, with 91.71% of responses verified by seven experts. Analysis of interaction logs revealed that medical questions significantly outnumbered logistical ones, hallucinations were negligible, and experts rated 84.52% of medical answers as accurate. As the knowledge base expanded with expert corrections, system performance improved by 19.02%, reducing expert workload. These insights guide the design of future LLM-powered chatbots., Comment: The first two authors contributed equally to this research

In the rapidly evolving field of vision-language navigation (VLN), ensuring robust safety mechanisms remains an open challenge. Control barrier functions (CBFs) are efficient tools which guarantee safety by solving an optimal control problem. In this work, we consider the case of a teleoperated drone in a VLN setting, and add safety features by formulating a novel scene-aware CBF using ego-centric observations obtained through an RGB-D sensor. As a baseline, we implement a vision-language understanding module which uses the contrastive language image pretraining (CLIP) model to query about a user-specified (in natural language) landmark. Using the YOLO (You Only Look Once) object detector, the CLIP model is queried for verifying the cropped landmark, triggering downstream navigation. To improve navigation safety of the baseline, we propose ASMA -- an Adaptive Safety Margin Algorithm -- that crops the drone's depth map for tracking moving object(s) to perform scene-aware CBF evaluation on-the-fly. By identifying potential risky observations from the scene, ASMA enables real-time adaptation to unpredictable environmental conditions, ensuring optimal safety bounds on a VLN-powered drone actions. Using the robot operating system (ROS) middleware on a parrot bebop2 quadrotor in the gazebo environment, ASMA offers 59.4% - 61.8% increase in success rates with insignificant 5.4% - 8.2% increases in trajectory lengths compared to the baseline CBF-less VLN while recovering from unsafe situations.

The ability of neural networks to perform robotic perception and control tasks such as depth and optical flow estimation, simultaneous localization and mapping (SLAM), and automatic control has led to their widespread adoption in recent years. Deep Reinforcement Learning has been used extensively in these settings, as it does not have the unsustainable training costs associated with supervised learning. However, DeepRL suffers from poor sample efficiency, i.e., it requires a large number of environmental interactions to converge to an acceptable solution. Modern RL algorithms such as Deep Q Learning and Soft Actor-Critic attempt to remedy this shortcoming but can not provide the explainability required in applications such as autonomous robotics. Humans intuitively understand the long-time-horizon sequential tasks common in robotics. Properly using such intuition can make RL policies more explainable while enhancing their sample efficiency. In this work, we propose SHIRE, a novel framework for encoding human intuition using Probabilistic Graphical Models (PGMs) and using it in the Deep RL training pipeline to enhance sample efficiency. Our framework achieves 25-78% sample efficiency gains across the environments we evaluate at negligible overhead cost. Additionally, by teaching RL agents the encoded elementary behavior, SHIRE enhances policy explainability. A real-world demonstration further highlights the efficacy of policies trained using our framework.

Organic spintronics has drawn the interest of the science community due to various applications in spin-valve devices. However, an efficient room-temperature Organic Spin Valve device has not been experimentally realized due to the complicated spin transport at the metal-organic interfaces. The present study focuses on a comprehensive understanding of the interfacial properties essential for advancing device performance and functionality. The structural and magnetic properties of the ultra-thin Cobalt (Co) films deposited on the fullerene (C60) layer are studied to investigate the origin of magnetic anisotropy in the metal-organic bilayer structures. Due to the mechanical softness of C60, penetration of ferromagnetic Co atoms inside the C60 film is confirmed by the X-ray reflectivity and Secondary Ion Mass Spectroscopy measurements. Grazing incidence small-angle X-ray scattering and atomic force microscopy provided information regarding the structural and morphological properties of the Co/C60 bilayers, angular dependent Magneto-optic Kerr effect measurements with varying Co layer thickness provided information about the growth-induced uniaxial magnetic anisotropy. In contrast to the inorganic silicon substrates, magnetic anisotropy in Co film tends to develop at 25 {\AA} thickness on the C60 layer, which further increases with the thickness of Cobalt. The anomalous behavior in coercivity and remanence variation along the nominal hard axis is explained by a two-grain Stoner-Wohlfarth model with intergranular exchange coupling. It is further confirmed by a non-uniform spatial distribution of magnetic domains investigated through Kerr microscopy. These anomalies could be attributed to the distribution of magneto-crystalline anisotropy and inhomogeneous strain caused by the formation of a diffused layer at the Co/C60 interface., Comment: 18 pages, 13 Figures

Learning semantically meaningful image transformations (i.e. rotation, thickness, blur) directly from examples can be a challenging task. Recently, the Manifold Autoencoder (MAE) proposed using a set of Lie group operators to learn image transformations directly from examples. However, this approach has limitations, as the learned operators are not guaranteed to be disentangled and the training routine is prohibitively expensive when scaling up the model. To address these limitations, we propose MANGO (transformation Manifolds with Grouped Operators) for learning disentangled operators that describe image transformations in distinct latent subspaces. Moreover, our approach allows practitioners the ability to define which transformations they aim to model, thus improving the semantic meaning of the learned operators. Through our experiments, we demonstrate that MANGO enables composition of image transformations and introduces a one-phase training routine that leads to a 100x speedup over prior works., Comment: Submitted to IEEE ICASSP 2025. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

Clinical notes contain unstructured representations of patient histories, including the relationships between medical problems and prescription drugs. To investigate the relationship between cancer drugs and their associated symptom burden, we extract structured, semantic representations of medical problem and drug information from the clinical narratives of oncology notes. We present Clinical Concept Annotations for Cancer Events and Relations (CACER), a novel corpus with fine-grained annotations for over 48,000 medical problems and drug events and 10,000 drug-problem and problem-problem relations. Leveraging CACER, we develop and evaluate transformer-based information extraction (IE) models such as BERT, Flan-T5, Llama3, and GPT-4 using fine-tuning and in-context learning (ICL). In event extraction, the fine-tuned BERT and Llama3 models achieved the highest performance at 88.2-88.0 F1, which is comparable to the inter-annotator agreement (IAA) of 88.4 F1. In relation extraction, the fine-tuned BERT, Flan-T5, and Llama3 achieved the highest performance at 61.8-65.3 F1. GPT-4 with ICL achieved the worst performance across both tasks. The fine-tuned models significantly outperformed GPT-4 in ICL, highlighting the importance of annotated training data and model optimization. Furthermore, the BERT models performed similarly to Llama3. For our task, LLMs offer no performance advantage over the smaller BERT models. The results emphasize the need for annotated training data to optimize models. Multiple fine-tuned transformer models achieved performance comparable to IAA for several extraction tasks., Comment: This is a pre-copy-editing, author-produced PDF of an article accepted for publication in JAMIA following peer review. The definitive publisher-authenticated version is available online at https://academic.oup.com/jamia/advance-article/doi/10.1093/jamia/ocae231/7748302

We initiate the study of relativistic zero-knowledge quantum proof of knowledge systems with classical communication, formally defining a number of useful concepts and constructing appropriate knowledge extractors for all the existing protocols in the relativistic setting which satisfy a weaker variant of the special soundness property due to Unruh (EUROCRYPT 2012). We show that there exists quantum proofs of knowledge with knowledge error 1/2 + negl({\eta}) for all relations in NP via a construction of such a system for the Hamiltonian cycle relation using a general relativistic commitment scheme exhibiting the fairly-binding property due to Fehr and Fillinger (EUROCRYPT 2016). We further show that one can construct quantum proof of knowledge extractors for proof systems which do not exhibit special soundness, and therefore require an extractor to rewind multiple times. We develop a new multi-prover quantum rewinding technique by combining ideas from monogamy of entanglement and gentle measurement lemmas that can break the quantum rewinding barrier. Finally, we prove a new bound on the impact of consecutive measurements and use it to significantly improve the soundness bound of some existing relativistic zero knowledge proof systems, such as the one due to Chailloux and Leverrier (EUROCRYPT 2017)., Comment: 37 pages

Invariants are essential for ensuring the security and correctness of Solidity smart contracts, particularly in the context of blockchain's immutability and decentralized execution. This paper introduces InvSol, a novel framework for pre-deployment invariant generation tailored specifically for Solidity smart contracts. Unlike existing solutions, namely InvCon, InvCon+, and Trace2Inv, that rely on post-deployment transaction histories on Ethereum mainnet, InvSol identifies invariants before deployment and offers comprehensive coverage of Solidity language constructs, including loops. Additionally, InvSol incorporates custom templates to effectively prevent critical issues such as reentrancy, out-of-gas errors, and exceptions during invariant generation. We rigorously evaluate InvSol using a benchmark set of smart contracts and compare its performance with state-of-the-art solutions. Our findings reveal that InvSol significantly outperforms these tools, demonstrating its effectiveness in handling new contracts with limited transaction histories. Notably, InvSol achieves a 15% improvement in identifying common vulnerabilities compared to InvCon+ and is able to address certain crucial vulnerabilities using specific invariant templates, better than Trace2Inv.

Excitation functions of several evaporation residues populated via complete and/or incomplete fusion in $^{12}$C+$^{193}$Ir system have been measured at energies $\approx$ 64-84 MeV, and analyzed in the framework of theoretical model code PACE4. It has been found that the $xn$ channels are predominantly populated via complete fusion; however, some of the $pxn$ channels decay via their precursor. A significant enhancement has been observed in the case of $\alpha$-emitting channels over PACE4 calculations, indicating the onset of a reaction mechanism not included in this code, e.g., incomplete fusion. For better insights into the onset and influence of incomplete fusion, the percentage fraction of incomplete fusion has been deduced and analyzed in terms of different entrance-channel parameters. The findings of the present study underline the importance of projectile energy, entrance-channel mass-asymmetry, and the Coulomb factor of interacting partners. The impact of projectile break-up on complete fusion has also been discussed in the framework of the Universal Fusion Function, where suppression of $\approx$ 12$\%$ has been observed in the fusion function. The finding of the present work reinstates that the fusion suppression is affected by the projectile $\alpha$-break-up threshold.

We explore the possibilities of modeling a spherically symmetric static spacetime that can emerge as the end state of gravitational collapse, by considering it to be seeded by a composite fluid made of matter and a scalar field. In this scenario, the matter represents dark matter, while the scalar field represents dark energy. On certain scales, dark energy is believed to significantly influence the structure formation of dark matter. Various models describe the possible impacts of dark energy on structure formation under different scenarios. By investigating an inhomogeneous scalar field representing dark energy, coupled with dark matter, we demonstrate that this two-component fluid can seed spacetimes forming the final equilibrium state. We derive solutions for the scalar field and potential for Joshi-Malafarina-Narayan (JMN) spacetimes., Comment: 15 pages, 3 figures

Raman scattering is an excellent method for simultaneously determining the dynamics of lattice, spin, and charge degrees of freedom. Furthermore, polarization selection rules in Raman scattering enable momentum-resolved quasiparticle dynamics. In this review, we highlight the potential of Raman scattering in probing magnetic quasiparticles or excitations in various magnetic materials. We demonstrate how temperature-dependent Raman scattering data can confirm the existence of magnons in long-range ordered magnets and fractionalized excitations in Kitaev spin liquid candidates. To make this review easily understandable to novices, we provide background information on magnons and fractionalized excitations, and explain how they become visible in the Raman scattering process. We also show how to estimate magnetic exchange interactions from the data. For both types of magnetic materials, we discuss the impact of spin-phonon coupling on the lineshape of the phonon modes. In terms of materials, we present magnetic Raman scattering data of antiferromagnetic Sr2IrO4 and La2CuO4, ferromagnetic CrI3 monolayers, and Kitaev spin liquid candidates {\alpha}-RuCl3 and \b{eta}-Li2IrO3. Overall, our review demonstrates the versatility of the Raman scattering technique in probing quasiparticles in magnetic quantum materials. The review aims to inform young experimental researchers about the potential of Raman scattering, thereby motivating them to use this technique in their research., Comment: It is a review article

Using a microscopic approach, we revisit the problem of superconducting critical temperature change in the presence of twin boundaries. We show that both critical temperature enhancement and suppression can come purely from geometric effects. These include aspects of scattering of electrons on these crystalline defects even when the coupling constant is unchanged. We consider two dimensional rectangular and three dimensional body centered cubic lattices with onsite s-wave superconducting pairing, nearest and next-to-nearest neighbor hoppings. In the considered two dimensional lattice with twin boundaries, the superconducting critical temperature associated with twinning planes is suppressed for moderate band filling and enhanced for an almost empty/filled band. The superconducting phase diagram is more diverse for the three dimensional lattice, which is caused by the interplay of van Hove singularity, changing coordination number, and modification of distances to nearest and next-to-nearest neighbors., Comment: 9 pages, 8 figures

The convergence of fully homomorphic encryption (FHE) and machine learning offers unprecedented opportunities for private inference of sensitive data. FHE enables computation directly on encrypted data, safeguarding the entire machine learning pipeline, including data and model confidentiality. However, existing FHE-based implementations for deep neural networks face significant challenges in computational cost, latency, and scalability, limiting their practical deployment. This paper introduces DCT-CryptoNets, a novel approach that leverages frequency-domain learning to tackle these issues. Our method operates directly in the frequency domain, utilizing the discrete cosine transform (DCT) commonly employed in JPEG compression. This approach is inherently compatible with remote computing services, where images are usually transmitted and stored in compressed formats. DCT-CryptoNets reduces the computational burden of homomorphic operations by focusing on perceptually relevant low-frequency components. This is demonstrated by substantial latency reduction of up to 5.3$\times$ compared to prior work on image classification tasks, including a novel demonstration of ImageNet inference within 2.5 hours, down from 12.5 hours compared to prior work on equivalent compute resources. Moreover, DCT-CryptoNets improves the reliability of encrypted accuracy by reducing variability (e.g., from $\pm$2.5\% to $\pm$1.0\% on ImageNet). This study demonstrates a promising avenue for achieving efficient and practical privacy-preserving deep learning on high resolution images seen in real-world applications., Comment: Under Review; 10 pages content, 3 pages appendix, 4 figures, 8 tables; Code TBD

In this work, we apply a phase-field modeling framework to elucidate the interplay between nucleation and kinetics in the dynamic evolution of twinning interfaces. The key feature of this phase-field approach is the ability to transparently and explicitly specify nucleation and kinetic behavior in the model, in contrast to other regularized interface models. We use this to study 2 distinct problems where it is essential to explicitly specify the kinetic and nucleation behavior governing twin evolution. First, we study twinning interfaces in 2-d. When these interfaces are driven to move, we find that significant levels of twin nucleation occur ahead of the moving interface. Essentially, the finite interface velocity and the relaxation time of the stresses ahead of the interface allows for nucleation to occur before the interface is able to propagate to that point. Second, we study the growth of needle twins in antiplane elasticity. We show that both nucleation and anisotropic kinetics are essential to obtain predictions of needle twins. While standard regularized interface approaches do not permit the transparent specification of anisotropic kinetics, this is readily possible with the phase-field approach that we have used here., Comment: To appear in Journal of Applied Mechanics

A comprehensive study of methods of maximizing Q for slow-wave coplanar waveguides is described. In addition to the widths of the signal conductor and coplanar ground lines and the distance between them, the length, spacing and stacking of the metal layers of the substrate shield strips are also shown to be critical in maximizing performance. Measured results from more than 50 different devices show that a 7X increase in the quality factor (e.g., Q > 70 at 24 GHz in 0.18 {\mu}m CMOS) is achievable using the optimum topology with optimum dimensions., Comment: 14 pages, 17 figures

Embedding Based Retrieval (EBR) is a crucial component of the retrieval stage in (Ads) Recommendation System that utilizes Two Tower or Siamese Networks to learn embeddings for both users and items (ads). It then employs an Approximate Nearest Neighbor Search (ANN) to efficiently retrieve the most relevant ads for a specific user. Despite the recent rise to popularity in the industry, they have a couple of limitations. Firstly, Two Tower model architecture uses a single dot product interaction which despite their efficiency fail to capture the data distribution in practice. Secondly, the centroid representation and cluster assignment, which are components of ANN, occur after the training process has been completed. As a result, they do not take into account the optimization criteria used for retrieval model. In this paper, we present Hierarchical Structured Neural Network (HSNN), a deployed jointly optimized hierarchical clustering and neural network model that can take advantage of sophisticated interactions and model architectures that are more common in the ranking stages while maintaining a sub-linear inference cost. We achieve 6.5% improvement in offline evaluation and also demonstrate 1.22% online gains through A/B experiments. HSNN has been successfully deployed into the Ads Recommendation system and is currently handling major portion of the traffic. The paper shares our experience in developing this system, dealing with challenges like freshness, volatility, cold start recommendations, cluster collapse and lessons deploying the model in a large scale retrieval production system., Comment: 9 pages

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

This paper explores humor detection through a linguistic lens, prioritizing syntactic, semantic, and contextual features over computational methods in Natural Language Processing. We categorize features into syntactic, semantic, and contextual dimensions, including lexicons, structural statistics, Word2Vec, WordNet, and phonetic style. Our proposed model, Colbert, utilizes BERT embeddings and parallel hidden layers to capture sentence congruity. By combining syntactic, semantic, and contextual features, we train Colbert for humor detection. Feature engineering examines essential syntactic and semantic features alongside BERT embeddings. SHAP interpretations and decision trees identify influential features, revealing that a holistic approach improves humor detection accuracy on unseen data. Integrating linguistic cues from different dimensions enhances the model's ability to understand humor complexity beyond traditional computational methods.

In memory computing (IMC) architectures for deep learning (DL) accelerators leverage energy-efficient and highly parallel matrix vector multiplication (MVM) operations, implemented directly in memory arrays. Such IMC designs have been explored based on CMOS as well as emerging non-volatile memory (NVM) technologies like RRAM. IMC architectures generally involve a large number of cores consisting of memory arrays, storing the trained weights of the DL model. Peripheral units like DACs and ADCs are also used for applying inputs and reading out the output values. Recently reported designs reveal that the ADCs required for reading out the MVM results, consume more than 85% of the total compute power and also dominate the area, thereby eschewing the benefits of the IMC scheme. Mitigation of imperfections in the ADCs, namely, non-linearity and variations, incur significant design overheads, due to dedicated calibration units. In this work we present peripheral aware design of IMC cores, to mitigate such overheads. It involves incorporating the non-idealities of ADCs in the training of the DL models, along with that of the memory units. The proposed approach applies equally well to both current mode as well as charge mode MVM operations demonstrated in recent years., and can significantly simplify the design of mixed-signal IMC units.

Large language models (LLMs) represent a groundbreaking advancement in the domain of natural language processing due to their impressive reasoning abilities. Recently, there has been considerable interest in increasing the context lengths for these models to enhance their applicability to complex tasks. However, at long context lengths and large batch sizes, the key-value (KV) cache, which stores the attention keys and values, emerges as the new bottleneck in memory usage during inference. To address this, we propose Eigen Attention, which performs the attention operation in a low-rank space, thereby reducing the KV cache memory overhead. Our proposed approach is orthogonal to existing KV cache compression techniques and can be used synergistically with them. Through extensive experiments over OPT, MPT, and Llama model families, we demonstrate that Eigen Attention results in up to 40% reduction in KV cache sizes and up to 60% reduction in attention operation latency with minimal drop in performance., Comment: 12 page, 6 figures, 6 tables

Given a set of co-located mobile robots in an unknown anonymous graph, the robots must relocate themselves in distinct graph nodes to solve the dispersion problem. In this paper, we consider the dispersion problem for silent robots \cite{gorain2024collaborative}, i.e., no direct, explicit communication between any two robots placed in the nodes of an oriented $n$ node ring network. The robots operate in synchronous rounds. The dispersion problem for silent mobile robots has been studied in arbitrary graphs where the robots start from a single source. In this paper, we focus on the dispersion problem for silent mobile robots where robots can start from multiple sources. The robots have unique labels from a range $[0,\;L]$ for some positive integer $L$. Any two co-located robots do not have the information about the label of the other robot. The robots have weak multiplicity detection capability, which means they can determine if it is alone on a node. The robots are assumed to be able to identify an increase or decrease in the number of robots present on a node in a particular round. However, the robots can not get the exact number of increase or decrease in the number of robots. We have proposed a deterministic distributed algorithm that solves the dispersion of $k$ robots in an oriented ring in $O(\log L+k)$ synchronous rounds with $O(\log L)$ bits of memory for each robot. A lower bound $\Omega(\log L+k)$ on time for the dispersion of $k$ robots on a ring network is presented to establish the optimality of the proposed algorithm.

Adversarial examples add imperceptible alterations to inputs with the objective to induce misclassification in machine learning models. They have been demonstrated to pose significant challenges in domains like image classification, with results showing that an adversarially perturbed image to evade detection against one classifier is most likely transferable to other classifiers. Adversarial examples have also been studied in malware analysis. Unlike images, program binaries cannot be arbitrarily perturbed without rendering them non-functional. Due to the difficulty of crafting adversarial program binaries, there is no consensus on the transferability of adversarially perturbed programs to different detectors. In this work, we explore the robustness of malware detectors against adversarially perturbed malware. We investigate the transferability of adversarial attacks developed against one detector, against other machine learning-based malware detectors, and code similarity techniques, specifically, locality sensitive hashing-based detectors. Our analysis reveals that adversarial program binaries crafted for one detector are generally less effective against others. We also evaluate an ensemble of detectors and show that they can potentially mitigate the impact of adversarial program binaries. Finally, we demonstrate that substantial program changes made to evade detection may result in the transformation technique being identified, implying that the adversary must make minimal changes to the program binary., Comment: This is the full version of the paper with the same title to appear in the proceedings of the 2024 Workshop on Security and Artificial Intelligence (SECAI 2024)

The frequency coexistence between geostationary orbit (GEO) and low earth orbit (LEO) satellite systems is expected to be a promising approach for relieving spectrum scarcity. However, it is essential to manage mutual interference between GEO and LEO satellite systems for frequency coexistence. Specifically, \emph{in-line interference}, caused by LEO satellites moving near the line-of-sight path between GEO satellite and GEO users (GUs), can significantly degrade GEO system throughput. This paper put forth a novel rate-splitting multiple access (RSMA) with a super-common message for GEO-LEO coexisting satellite systems (CSS). By employing a super-common message that GUs can decode, GUs can mitigate the in-line interference by successive interference cancellation (SIC). Moreover, we formulate a traffic-aware throughput maximization (TTM) problem to satisfy the heterogeneous traffic demands of users by minimizing total unmet throughput demands (or user dissatisfaction). By doing so, the TTM precoder can be flexibly adjusted according to the interference leakage from LEO satellites to GUs and target traffic demands. Numerical results confirm that our proposed method ensures seamless connectivity even in the GEO-LEO in-line interference regime under imperfect channel state information (CSI) at both the transmitter and receiver., Comment: 17 pages, 4 figures, 1 table

This paper introduces a Scalable Hierarchical Aware Convolutional Neural Network (SHA-CNN) model architecture for Edge AI applications. The proposed hierarchical CNN model is meticulously crafted to strike a balance between computational efficiency and accuracy, addressing the challenges posed by resource-constrained edge devices. SHA-CNN demonstrates its efficacy by achieving accuracy comparable to state-of-the-art hierarchical models while outperforming baseline models in accuracy metrics. The key innovation lies in the model's hierarchical awareness, enabling it to discern and prioritize relevant features at multiple levels of abstraction. The proposed architecture classifies data in a hierarchical manner, facilitating a nuanced understanding of complex features within the datasets. Moreover, SHA-CNN exhibits a remarkable capacity for scalability, allowing for the seamless incorporation of new classes. This flexibility is particularly advantageous in dynamic environments where the model needs to adapt to evolving datasets and accommodate additional classes without the need for extensive retraining. Testing has been conducted on the PYNQ Z2 FPGA board to validate the proposed model. The results achieved an accuracy of 99.34%, 83.35%, and 63.66% for MNIST, CIFAR-10, and CIFAR-100 datasets, respectively. For CIFAR-100, our proposed architecture performs hierarchical classification with 10% reduced computation while compromising only 0.7% accuracy with the state-of-the-art. The adaptability of SHA-CNN to FPGA architecture underscores its potential for deployment in edge devices, where computational resources are limited. The SHA-CNN framework thus emerges as a promising advancement in the intersection of hierarchical CNNs, scalability, and FPGA-based Edge AI.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

eBay partners with external vendors, which allows customers to freely select a vendor to complete their eBay experiences. However, vendor outages can hinder customer experiences. Consequently, eBay can disable a problematic vendor to prevent customer loss. Disabling the vendor too late risks losing customers willing to switch to other vendors, while disabling it too early risks losing those unwilling to switch. In this paper, we propose a data-driven solution to answer whether eBay should disable a problematic vendor and when to disable it. Our solution involves forecasting customer behavior. First, we use a multiplicative seasonality model to represent behavior if all vendors are fully functioning. Next, we use a Monte Carlo simulation to represent behavior if the problematic vendor remains enabled. Finally, we use a linear model to represent behavior if the vendor is disabled. By comparing these forecasts, we determine the optimal time for eBay to disable the problematic vendor.

We discuss the future prospects of a minimally five-dimensional version of the well-motivated scenario for addressing the discrepancy in the muon anomalous magnetic moment, the $U(1)_{L_\mu - L_\tau}$ extension of the standard model (SM) gauge symmetry. Here, multiple associated massive gauge bosons appear thanks to the five-dimensional $U(1)_{L_\mu - L_\tau}$ gauge symmetry, and they contribute to the muon $(g-2)$ and also other processes. We focus on the powerful probe of elastic neutrino-electron scatterings since the upcoming DUNE experiment will explore MeV-scale uncharted regions by previous experiments (e.g., CHARM-II and Borexino) in the near future. We found that even with small kinetic mixing parameters, much of the parameter space, including those satisfying muon $(g-2)$, can be probed using several years of data from the DUNE experiment, focusing on the near detector. In our scenario, interference effects between intermediate-state gauge bosons play an important role. Our results include comparisons between flat and warped extra dimensions., Comment: 40 pages, 7 figures

The advent of novel high-entropy oxides has sparked substantial research interest due to their exceptional functional properties, which often surpass the mere sum of their constituent elements' characteristics. This study introduces a complex high-entropy spinel oxide with composition (Ni$_{0.2}$Mg$_{0.2}$Co$_{0.2}$Cu$_{0.2}$Zn$_{0.2}$)(Mn$_{0.66}$Fe$_{0.66}$Cr$_{0.66}$)O$_{4}$. We performed comprehensive structural (X-ray and Neutron diffraction), microstructural, magnetic, and local electronic structure investigations on this material. Despite the material's high degree of disorder, detailed magnetization measurements and low temperature neutron powder diffraction studies reveal long-range ferrimagnetic ordering beginning at 293 K. The sample exhibits a high saturation magnetization of 766 emu-cm${^3}$ (at 50 K), a low coercivity (H$_C$) of 100 Oe (50 K), a high transition temperature (T$_C$) around room temperature, and high resistivity value of 4000 Ohm-cm at room temperature, indicating its potential for high density memory devices. The magnetic structure is determined using a collinear-type ferrimagnetic model with a propagation vector k = 0,0,0. Various analytical techniques, including modified Arrott plots, Kouvel-Fischer analysis, and critical isotherm analysis, are employed to investigate the phase transitions and magnetic properties of this complex system. Our results indicate a second-order phase transition. Remarkably, despite the complex structure and significant disorder, the critical exponents obtained are consistent with the mean field model. The high entropy leads to a remarkably homogeneous distribution of multiple cations, validating the approximation of average local magnetic environments and supporting the mean field theory.

Generative AI, especially via Large Language Models (LLMs), has transformed content creation across text, images, and music, showcasing capabilities in following instructions through prompting, largely facilitated by instruction tuning. Instruction tuning is a supervised fine-tuning method where LLMs are trained on datasets formatted with specific tasks and corresponding instructions. This method systematically enhances the model's ability to comprehend and execute the provided directives. Despite these advancements, LLMs still face challenges in consistently interpreting complex, multi-step instructions and generalizing them to novel tasks, which are essential for broader applicability in real-world scenarios. This article explores why neurosymbolic AI offers a better path to enhance the instructability of LLMs. We explore the use a symbolic task planner to decompose high-level instructions into structured tasks, a neural semantic parser to ground these tasks into executable actions, and a neuro-symbolic executor to implement these actions while dynamically maintaining an explicit representation of state. We also seek to show that neurosymbolic approach enhances the reliability and context-awareness of task execution, enabling LLMs to dynamically interpret and respond to a wider range of instructional contexts with greater precision and flexibility.

The recent advent of 3D Gaussian Splatting (3DGS) has revolutionized the 3D scene reconstruction space enabling high-fidelity novel view synthesis in real-time. However, with the exception of RawNeRF, all prior 3DGS and NeRF-based methods rely on 8-bit tone-mapped Low Dynamic Range (LDR) images for scene reconstruction. Such methods struggle to achieve accurate reconstructions in scenes that require a higher dynamic range. Examples include scenes captured in nighttime or poorly lit indoor spaces having a low signal-to-noise ratio, as well as daylight scenes with shadow regions exhibiting extreme contrast. Our proposed method HDRSplat tailors 3DGS to train directly on 14-bit linear raw images in near darkness which preserves the scenes' full dynamic range and content. Our key contributions are two-fold: Firstly, we propose a linear HDR space-suited loss that effectively extracts scene information from noisy dark regions and nearly saturated bright regions simultaneously, while also handling view-dependent colors without increasing the degree of spherical harmonics. Secondly, through careful rasterization tuning, we implicitly overcome the heavy reliance and sensitivity of 3DGS on point cloud initialization. This is critical for accurate reconstruction in regions of low texture, high depth of field, and low illumination. HDRSplat is the fastest method to date that does 14-bit (HDR) 3D scene reconstruction in $\le$15 minutes/scene ($\sim$30x faster than prior state-of-the-art RawNeRF). It also boasts the fastest inference speed at $\ge$120fps. We further demonstrate the applicability of our HDR scene reconstruction by showcasing various applications like synthetic defocus, dense depth map extraction, and post-capture control of exposure, tone-mapping and view-point.

In a few recent manuscripts, we used the affine connection to introduce two massless scalar fields in the Einstein-Palatini action. These fields lead to non-metricity. In this article, we will discuss the significance of these fields in inflation and dark energy. We will construct a Lagrangian formalism to include these scalar fields in a theory of gravity coupled with ordinary matter and radiation. We will find that these fields need not to be included in the actions of interacting gauge theories coupled with conserved fermionic vector currents as a part of the connection. The same remains valid for ordinary scalar fields. We can couple the connection-scalars with ordinary matter by adding suitable interaction terms. In this context, we will find that Stokes' theorem leads us to include the right-handed neutrinos in the electroweak theory in curved spacetime even with the Levi-Civita connection. This is required to obtain consistent equations of motion and anomaly-free conserved vector current for the neutrinos. Axial vector currents for different Dirac fields may remain anomalous. The right-handed neutrinos can be useful to explain neutrino oscillation and dark matter. We will also discuss the possibility of introducing massless finite integer spin particles using second rank symmetric traceless tensors with reference to the corresponding little group in flat spacetime. We will show that we can use massless (A,A) type fields in Minkowski space to introduce massless finite integer spin particles., Comment: 17 pages, accepted in Quantum Studies: Mathematics and Foundations

We present results from a search for X-ray/gamma-ray counterparts of gravitational-wave (GW) candidates from the third observing run (O3) of the LIGO-Virgo-KAGRA (LVK) network using the Swift Burst Alert Telescope (Swift-BAT). The search includes 636 GW candidates received in low latency, 86 of which have been confirmed by the offline analysis and included in the third cumulative Gravitational-Wave Transient Catalogs (GWTC-3). Targeted searches were carried out on the entire GW sample using the maximum--likelihood NITRATES pipeline on the BAT data made available via the GUANO infrastructure. We do not detect any significant electromagnetic emission that is temporally and spatially coincident with any of the GW candidates. We report flux upper limits in the 15-350 keV band as a function of sky position for all the catalog candidates. For GW candidates where the Swift-BAT false alarm rate is less than 10$^{-3}$ Hz, we compute the GW--BAT joint false alarm rate. Finally, the derived Swift-BAT upper limits are used to infer constraints on the putative electromagnetic emission associated with binary black hole mergers., Comment: 50 pages, 10 figures, 4 tables

Different from human nature, it is still common practice today for vision tasks to train deep learning models only initially and on fixed datasets. A variety of approaches have recently addressed handling continual data streams. However, extending these methods to manage out-of-distribution (OOD) scenarios has not effectively been investigated. On the other hand, it has recently been shown that non-continual neural mesh models exhibit strong performance in generalizing to such OOD scenarios. To leverage this decisive property in a continual learning setting, we propose incremental neural mesh models that can be extended with new meshes over time. In addition, we present a latent space initialization strategy that enables us to allocate feature space for future unseen classes in advance and a positional regularization term that forces the features of the different classes to consistently stay in respective latent space regions. We demonstrate the effectiveness of our method through extensive experiments on the Pascal3D and ObjectNet3D datasets and show that our approach outperforms the baselines for classification by $2-6\%$ in the in-domain and by $6-50\%$ in the OOD setting. Our work also presents the first incremental learning approach for pose estimation. Our code and model can be found at https://github.com/Fischer-Tom/iNeMo., Comment: ECCV-24

In this article, we propose a novel standalone hybrid Spiking-Convolutional Neural Network (SC-NN) model and test on using image inpainting tasks. Our approach uses the unique capabilities of SNNs, such as event-based computation and temporal processing, along with the strong representation learning abilities of CNNs, to generate high-quality inpainted images. The model is trained on a custom dataset specifically designed for image inpainting, where missing regions are created using masks. The hybrid model consists of SNNConv2d layers and traditional CNN layers. The SNNConv2d layers implement the leaky integrate-and-fire (LIF) neuron model, capturing spiking behavior, while the CNN layers capture spatial features. In this study, a mean squared error (MSE) loss function demonstrates the training process, where a training loss value of 0.015, indicates accurate performance on the training set and the model achieved a validation loss value as low as 0.0017 on the testing set. Furthermore, extensive experimental results demonstrate state-of-the-art performance, showcasing the potential of integrating temporal dynamics and feature extraction in a single network for image inpainting., Comment: 7 Pages, 3 figures, and 2 tables

Neural rendering methods can achieve near-photorealistic image synthesis of scenes from posed input images. However, when the images are imperfect, e.g., captured in very low-light conditions, state-of-the-art methods fail to reconstruct high-quality 3D scenes. Recent approaches have tried to address this limitation by modeling various degradation processes in the image formation model; however, this limits them to specific image degradations. In this paper, we propose a generalizable neural rendering method that can perform high-fidelity novel view synthesis under several degradations. Our method, GAURA, is learning-based and does not require any test-time scene-specific optimization. It is trained on a synthetic dataset that includes several degradation types. GAURA outperforms state-of-the-art methods on several benchmarks for low-light enhancement, dehazing, deraining, and on-par for motion deblurring. Further, our model can be efficiently fine-tuned to any new incoming degradation using minimal data. We thus demonstrate adaptation results on two unseen degradations, desnowing and removing defocus blur. Code and video results are available at vinayak-vg.github.io/GAURA., Comment: European Conference on Computer Vision(ECCV) 2024

Multi-robot collaboration has become a needed component in unknown environment exploration due to its ability to accomplish various challenging situations. Potential-field-based methods are widely used for autonomous exploration because of their high efficiency and low travel cost. However, exploration speed and collaboration ability are still challenging topics. Therefore, we propose a Distributed Multi-Robot Potential-Field-Based Exploration (DMPF-Explore). In particular, we first present a Distributed Submap-Based Multi-Robot Collaborative Mapping Method (DSMC-Map), which can efficiently estimate the robot trajectories and construct the global map by merging the local maps from each robot. Second, we introduce a Potential-Field-Based Exploration Strategy Augmented with Modified Wave-Front Distance and Colored Noises (MWF-CN), in which the accessible frontier neighborhood is extended, and the colored noise provokes the enhancement of exploration performance. The proposed exploration method is deployed for simulation and real-world scenarios. The results show that our approach outperforms the existing ones regarding exploration speed and collaboration ability., Comment: This paper has been accepted by Robotics and Autonomous Systems

Chronotype compares individuals' circadian phase to others. It contextualizes mental health risk assessments and detection of social jet lag, which can hamper mental health and cognitive performance. Existing ways of determining chronotypes, such as Dim Light Melatonin Onset (DLMO) or the Morningness-Eveningness Questionnaire (MEQ), are limited by being discrete in time and time-intensive to update, meaning they rarely capture real-world variability across time. Chronotyping users based on a daily planner app might augment existing methods to enable assessment continuously and at scale. This paper reports the construction of a supervised binary classifier that attempts to demonstrate the feasibility of this approach. 1,460 registered users from the Owaves app opted in by filling out the MEQ survey between July 14, 2022, and May 1, 2023. 142 met the eligibility criteria. We used multimodal app data from individuals identified as morning and evening types from MEQ data, basing the classifier on app time series data. This included daily timing for 8 main lifestyle activity types: exercise, sleep, social interactions, meal times, relaxation, work, play, and miscellaneous, as defined in the app. The timing of activities showed substantial change across time, as well as heterogeneity by activity type. Our novel chronotyping classifier was able to predict the morningness and eveningness of its users with an ROC AUC of 0.70. Our findings demonstrate the feasibility of chronotype classification from multimodal, real-world app data, while highlighting fundamental challenges to applying discrete and fixed labels to complex, dynamic, multimodal behaviors. Our findings suggest a potential for real-time monitoring of shifts in chronotype specific to different causes (i.e. types of activity), which could feasibly be used to support future, prospective mental health support research., Comment: 15 pages, 4 figures, unsubmitted for peer review at date of posting; New Version 081724: expanded results text and updated discussion to match. Minor changes for clarity throughout

Graph games lie at the algorithmic core of many automated design problems in computer science. These are games usually played between two players on a given graph, where the players keep moving a token along the edges according to pre-determined rules, and the winner is decided based on the infinite path traversed by the token from a given initial position. In bidding games, the players initially get some monetary budgets which they need to use to bid for the privilege of moving the token at each step. Each round of bidding affects the players' available budgets, which is the only form of update that the budgets experience. We introduce bidding games with charging where the players can additionally improve their budgets during the game by collecting vertex-dependent charges. Unlike traditional bidding games (where all charges are zero), bidding games with charging allow non-trivial recurrent behaviors. We show that the central property of traditional bidding games generalizes to bidding games with charging: For each vertex there exists a threshold ratio, which is the necessary and sufficient fraction of the total budget for winning the game from that vertex. While the thresholds of traditional bidding games correspond to unique fixed points of linear systems of equations, in games with charging, these fixed points are no longer unique. This significantly complicates the proof of existence and the algorithmic computation of thresholds for infinite-duration objectives. We also provide the lower complexity bounds for computing thresholds for Rabin and Streett objectives, which are the first known lower bounds in any form of bidding games (with or without charging), and we solve the following repair problem for safety and reachability games that have unsatisfiable objectives: Can we distribute a given amount of charge to the players in a way such that the objective can be satisfied?