Search

In this paper, we propose a network architecture where two types of aerial infrastructures together with a ground station provide connectivity to a remote area. A high altitude platform station (HAPS) is equipped with reconfigurable intelligent surface (RIS), so-called HAPS-RIS, to be exploited to assist the unmanned aerial vehicle (UAV)-based wireless networks. A key challenge in such networks is the restricted number of UAVs, which limits full coverage and leaves some users unsupported. To tackle this issue, we propose a hierarchical bilevel optimization framework including a leader and a follower problem. The users served by HAPS-RIS are in a zone called HAPS-RIS zone and the users served by the UAVs are in another zone called UAV zone. In the leader problem, the goal is to establish the zone boundary that maximizes the number of users covered by HAPS-RIS while ensuring that users in this zone meet their rate requirements. This is achieved through an algorithm that integrates RIS clustering, subcarrier allocation, and zone determination. The follower problem focuses on minimizing the number of UAVs required, ensuring that the rate requirements of the users in the UAV zone are met. This is addressed using an algorithm that employs k-means clustering and subcarrier allocation. Our study reveals that increasing the number of RIS elements significantly decreases the number of required UAVs.

In smart manufacturing, scheduling autonomous internal logistic vehicles is crucial for optimizing operational efficiency. This paper proposes a multi-agent deep Q-network (MADQN) with a layer-based communication channel (LBCC) to address this challenge. The main goals are to minimize total job tardiness, reduce the number of tardy jobs, and lower vehicle energy consumption. The method is evaluated against nine well-known scheduling heuristics, demonstrating its effectiveness in handling dynamic job shop behaviors like job arrivals and workstation unavailabilities. The approach also proves scalable, maintaining performance across different layouts and larger problem instances, highlighting the robustness and adaptability of MADQN with LBCC in smart manufacturing., Comment: Accepted for the 5th IFAC/INSTICC INTERNATIONAL CONFERENCE ON INNOVATIVE INTELLIGENT INDUSTRIAL PRODUCTION AND LOGISTICS

Multi-document (MD) processing is crucial for LLMs to handle real-world tasks such as summarization and question-answering across large sets of documents. While LLMs have improved at processing long inputs, MD contexts still present challenges, such as managing inter-document dependencies, redundancy, and incoherent structures. We introduce MDCure, a scalable and effective fine-tuning pipeline to enhance the MD capabilities of LLMs without the computational cost of pre-training or reliance on human annotated data. MDCure is based on generation of high-quality synthetic MD instruction data from sets of related articles via targeted prompts. We further introduce MDCureRM, a multi-objective reward model which filters generated data based on their training utility for MD settings. With MDCure, we fine-tune a variety of LLMs, from the FlanT5, Qwen2, and LLAMA3.1 model families, up to 70B parameters in size. Extensive evaluations on a wide range of MD and long-context benchmarks spanning various tasks show MDCure consistently improves performance over pre-trained baselines and over corresponding base models by up to 75.5%. Our code, datasets, and models are available at https://github.com/yale-nlp/MDCure.

Existing benchmarks often highlight the remarkable performance achieved by state-of-the-art Multimodal Foundation Models (MFMs) in leveraging temporal context for video understanding. However, how well do the models truly perform visual temporal reasoning? Our study of existing benchmarks shows that this capability of MFMs is likely overestimated as many questions can be solved by using a single, few, or out-of-order frames. To systematically examine current visual temporal reasoning tasks, we propose three principles with corresponding metrics: (1) Multi-Frame Gain, (2) Frame Order Sensitivity, and (3) Frame Information Disparity. Following these principles, we introduce TOMATO, Temporal Reasoning Multimodal Evaluation, a novel benchmark crafted to rigorously assess MFMs' temporal reasoning capabilities in video understanding. TOMATO comprises 1,484 carefully curated, human-annotated questions spanning six tasks (i.e., action count, direction, rotation, shape & trend, velocity & frequency, and visual cues), applied to 1,417 videos, including 805 self-recorded and -generated videos, that encompass human-centric, real-world, and simulated scenarios. Our comprehensive evaluation reveals a human-model performance gap of 57.3% with the best-performing model. Moreover, our in-depth analysis uncovers more fundamental limitations beyond this gap in current MFMs. While they can accurately recognize events in isolated frames, they fail to interpret these frames as a continuous sequence. We believe TOMATO will serve as a crucial testbed for evaluating the next-generation MFMs and as a call to the community to develop AI systems capable of comprehending human world dynamics through the video modality.

Many alignment methods, including reinforcement learning from human feedback (RLHF), rely on the Bradley-Terry reward assumption, which is insufficient to capture the full range of general human preferences. To achieve robust alignment with general preferences, we model the alignment problem as a two-player zero-sum game, where the Nash equilibrium policy guarantees a 50% win rate against any competing policy. However, previous algorithms for finding the Nash policy either diverge or converge to a Nash policy in a modified game, even in a simple synthetic setting, thereby failing to maintain the 50% win rate guarantee against all other policies. We propose a meta-algorithm, Convergent Meta Alignment Algorithm (COMAL), for language model alignment with general preferences, inspired by convergent algorithms in game theory. Theoretically, we prove that our meta-algorithm converges to an exact Nash policy in the last iterate. Additionally, our meta-algorithm is simple and can be integrated with many existing methods designed for RLHF and preference optimization with minimal changes. Experimental results demonstrate the effectiveness of the proposed framework when combined with existing preference policy optimization methods.

Like many chronic diseases, human immunodeficiency virus (HIV) is managed over time at regular clinic visits. At each visit, patient features are assessed, treatments are prescribed, and a subsequent visit is scheduled. There is a need for data-driven methods for both predicting retention and recommending scheduling decisions that optimize retention. Prediction models can be useful for estimating retention rates across a range of scheduling options. However, training such models with electronic health records (EHR) involves several complexities. First, formal causal inference methods are needed to adjust for observed confounding when estimating retention rates under counterfactual scheduling decisions. Second, competing events such as death preclude retention, while censoring events render retention missing. Third, inconsistent monitoring of features such as viral load and CD4 count lead to covariate missingness. This paper presents an all-in-one approach for both predicting HIV retention and optimizing scheduling while accounting for these complexities. We formulate and identify causal retention estimands in terms of potential return-time under a hypothetical scheduling decision. Flexible Bayesian approaches are used to model the observed return-time distribution while accounting for competing and censoring events and form posterior point and uncertainty estimates for these estimands. We address the urgent need for data-driven decision support in HIV care by applying our method to EHR from the Academic Model Providing Access to Healthcare (AMPATH) - a consortium of clinics that treat HIV in Western Kenya.

Travel planning is a challenging and time-consuming task that aims to find an itinerary which satisfies multiple, interdependent constraints regarding flights, accommodations, attractions, and other travel arrangements. In this paper, we propose To the Globe (TTG), a real-time demo system that takes natural language requests from users, translates it to symbolic form via a fine-tuned Large Language Model, and produces optimal travel itineraries with Mixed Integer Linear Programming solvers. The overall system takes ~5 seconds to reply to the user request with guaranteed itineraries. To train TTG, we develop a synthetic data pipeline that generates user requests, flight and hotel information in symbolic form without human annotations, based on the statistics of real-world datasets, and fine-tune an LLM to translate NL user requests to their symbolic form, which is sent to the symbolic solver to compute optimal itineraries. Our NL-symbolic translation achieves ~91% exact match in a backtranslation metric (i.e., whether the estimated symbolic form of generated natural language matches the groundtruth), and its returned itineraries have a ratio of 0.979 compared to the optimal cost of the ground truth user request. When evaluated by users, TTG achieves consistently high Net Promoter Scores (NPS) of 35-40% on generated itinerary.

This paper addresses the problem of detecting changes when only unnormalized pre- and post-change distributions are accessible. This situation happens in many scenarios in physics such as in ferromagnetism, crystallography, magneto-hydrodynamics, and thermodynamics, where the energy models are difficult to normalize. Our approach is based on the estimation of the Cumulative Sum (CUSUM) statistics, which is known to produce optimal performance. We first present an intuitively appealing approximation method. Unfortunately, this produces a biased estimator of the CUSUM statistics and may cause performance degradation. We then propose the Log-Partition Approximation Cumulative Sum (LPA-CUSUM) algorithm based on thermodynamic integration (TI) in order to estimate the log-ratio of normalizing constants of pre- and post-change distributions. It is proved that this approach gives an unbiased estimate of the log-partition function and the CUSUM statistics, and leads to an asymptotically optimal performance. Moreover, we derive a relationship between the required sample size for thermodynamic integration and the desired detection delay performance, offering guidelines for practical parameter selection. Numerical studies are provided demonstrating the efficacy of our approach.

Deep learning has achieved great success in modeling dynamical systems, providing data-driven simulators to predict complex phenomena, even without known governing equations. However, existing models have two major limitations: their narrow focus on mechanical systems and their tendency to treat systems as monolithic. These limitations reduce their applicability to dynamical systems in other domains, such as electrical and hydraulic systems, and to coupled systems. To address these limitations, we propose Poisson-Dirac Neural Networks (PoDiNNs), a novel framework based on the Dirac structure that unifies the port-Hamiltonian and Poisson formulations from geometric mechanics. This framework enables a unified representation of various dynamical systems across multiple domains as well as their interactions and degeneracies arising from couplings. Our experiments demonstrate that PoDiNNs offer improved accuracy and interpretability in modeling unknown coupled dynamical systems from data.

We investigate the ionization state of very energetic cosmic ray nuclei in photon fields, such as the cosmic microwave background (CMB) in extragalactic propagation and the environment surrounding the acceleration site in astrophysical sources. We focus on the process of pair production with electron capture (PPC), where the interaction of a single photon with a nucleus produces an $e^\pm$ pair (similar to Bethe-Heitler process) with the subsequent capture of electron by the nucleus. This process effectively reduces the nucleus charge by one unit and counteracts the photo-ionization process. We show that during cosmological propagation, where the ultra-high energy cosmic rays interact predominantly with the CMB, ionization dominates over PPC for all the cases of practical interest. However, within the source environment and at sufficiently high energies, ionization and PPC processes can reach an equilibrium, leading to a significant fraction of dressed heavy nuclei. This provides a further limitation to the acceleration of high-$Z$ nuclei to very high energies in environments with hot and dense photon fields., Comment: 11 pages, 5 figures, 1 table

Orthogonal time frequency space modulation (OTFS) is currently one of the most robust modulation techniques for high Doppler channels. However, to reap the benefits of OTFS, an accurate channel estimation is crucial. To this mean, the widely used embedded pilot structures use twice the channel length size as a delay guard to avoid interference between the pilot and data symbols. Hence, incurring a large spectral efficiency loss, especially in wideband systems where the channel length is large. To reduce the pilot overhead, we propose a novel split pilot structure with two impulse pilots. With two pilots, we can use one to cancel the other, thus, capable of removing the pilot interference over data. To remove the data interference from the pilot, we also propose an iterative joint channel estimation and detection technique tailored to the proposed split pilot structure. With the interference caused by the delay spread solved, we reduce the number of delay guards in our system by half, significantly improving the spectral efficiency. To corroborate our claims, we numerically demonstrate that our proposed method can achieve performance levels comparable to that of the full-guard method while using only half the delay guard. Additionally, we show that our proposed iterative channel estimating technique has a fast convergence speed, requiring only two iterations., Comment: Accepted in Globecom workshops 2024

In this paper, we propose time and frequency synchronization techniques for the uplink of multiuser OTFS (MU-OTFS) in high-mobility scenarios. We introduce a spectrally efficient and practical pilot pattern where each user utilizes a pilot with a cyclic prefix (PCP) within a shared pilot region on the delay-Doppler plane. At the receiver, a bank of filters is deployed to separate the users' signals and accurately estimate their timing offsets (TOs) and carrier frequency offsets (CFOs). Our technique employs a threshold-based approach that provides precise TO estimates. Our proposed CFO estimation technique reduces the multi-dimensional maximum likelihood (ML) search problem into multiple one-dimensional search problems. Furthermore, we apply the Chebyshev polynomials of the first kind basis expansion model (CPF-BEM) to effectively handle the time-variations of the channel in obtaining the CFO estimates for all the users. Finally, we numerically investigate the error performance of our proposed synchronization technique in high mobility scenarios for the MU-OTFS uplink. Our simulation results confirm the efficacy of the proposed technique in estimating the TOs and CFOs which also leads to an improved channel estimation performance.

Existing methods on understanding the capabilities of LLMs in logical reasoning rely on binary entailment classification or synthetically derived rationales, which are not sufficient for proper investigation of model's capabilities. We present P-FOLIO, a human-annotated dataset consisting of diverse and complex reasoning chains for a set of realistic logical reasoning stories also written by humans. P-FOLIO is collected with an annotation protocol that facilitates humans to annotate well-structured natural language proofs for first-order logic reasoning problems in a step-by-step manner. The number of reasoning steps in P-FOLIO span from 0 to 20. We further use P-FOLIO to evaluate and improve large-language-model (LLM) reasoning capabilities. We evaluate LLM reasoning capabilities at a fine granularity via single-step inference rule classification, with more diverse inference rules of more diverse and higher levels of complexities than previous works. Given that a single model-generated reasoning chain could take a completely different path than the human-annotated one, we sample multiple reasoning chains from a model and use pass@k metrics for evaluating the quality of model-generated reasoning chains. We show that human-written reasoning chains significantly boost the logical reasoning capabilities of LLMs via many-shot prompting and fine-tuning. Furthermore, fine-tuning Llama3-7B on P-FOLIO improves the model performance by 10% or more on three other out-of-domain logical reasoning datasets. We also conduct detailed analysis to show where most powerful LLMs fall short in reasoning. We will release the dataset and code publicly.

Hydrogen production is a vital process in the quest for decarbonization and a sustainable future. This conversation explores the various technologies used in hydrogen production, such as steam methane reforming, electrolysis, and biomass gasification. These methods have different applications and efficiencies, but all contribute to the production of hydrogen gas. The future of hydrogen production looks promising, as it offers a clean and versatile energy source that can be used in various sectors, including transportation, industry, and power generation. The increasing demand for hydrogen in these sectors, coupled with the global push for decarbonization, highlights the importance of advancing hydrogen production technologies and infrastructure. This paper focuses broadly on different methods of hydrogen production like steam methane reforming, electrolysis, and biomass gasification. Steam methane reforming is currently the most common method, accounting for about 95% of global hydrogen production. Electrolysis is another method that uses electricity to split water into hydrogen and oxygen. Biomass gasification involves converting organic materials into hydrogen gas through a thermochemical process. The renewable energy sources considered are water and biomass and the methods considered are Electrolysis (grid), thermolysis, thermochemical water splitting, photoelectrochemical water splitting, and gasification. Electrolysis of water to produce hydrogen accounts for about 5% of the total hydrogen production. Statistically, global hydrogen production reached about USD 155.35 billion in 2022, with the majority of it being used in the petroleum refining and ammonia production industries. As for applications, hydrogen can be used as a fuel for fuel cell vehicles, as a feedstock for chemical processes, and as a storage medium for renewable energy., Comment: 8 pages, 8 figures, International Conference on Advances in Chemical and Applied Sciences for Sustainable Development (ACASSD-2024)

The automatic evaluation of instruction following typically involves using large language models (LLMs) to assess response quality. However, there is a lack of comprehensive evaluation of these LLM-based evaluators across two dimensions: the base LLMs and the evaluation protocols. Therefore, we present a thorough meta-evaluation of instruction following, including 25 base LLMs and 15 recently proposed evaluation protocols, on 4 human-annotated datasets, assessing the evaluation accuracy of the LLM-evaluators. Our evaluation allows us to identify the best-performing base LLMs and evaluation protocols with a high degree of robustness. Moreover, our large-scale evaluation reveals: (1) Base LLM performance ranking remains largely consistent across evaluation protocols, with less capable LLMs showing greater improvement from protocol enhancements; (2) Robust evaluation of evaluation protocols requires many base LLMs with varying capability levels, as protocol effectiveness can depend on the base LLM used; (3) Evaluation results on different datasets are not always consistent, so a rigorous evaluation requires multiple datasets with distinctive features. We release our meta-evaluation suite ReIFE, which provides the codebase and evaluation result collection for more than 500 LLM-evaluator configurations, to support future research in instruction-following evaluation., Comment: GitHub Repo: https://github.com/yale-nlp/ReIFE, Evaluation Result Collection: https://huggingface.co/datasets/yale-nlp/ReIFE

Large language models (LLMs) are becoming increasingly prevalent in modern software systems, interfacing between the user and the internet to assist with tasks that require advanced language understanding. To accomplish these tasks, the LLM often uses external data sources such as user documents, web retrieval, results from API calls, etc. This opens up new avenues for attackers to manipulate the LLM via prompt injection. Adversarial prompts can be carefully crafted and injected into external data sources to override the user's intended instruction and instead execute a malicious instruction. Prompt injection attacks constitute a major threat to LLM security, making the design and implementation of practical countermeasures of paramount importance. To this end, we show that alignment can be a powerful tool to make LLMs more robust against prompt injection. Our method -- SecAlign -- first builds an alignment dataset by simulating prompt injection attacks and constructing pairs of desirable and undesirable responses. Then, we apply existing alignment techniques to fine-tune the LLM to be robust against these simulated attacks. Our experiments show that SecAlign robustifies the LLM substantially with a negligible hurt on model utility. Moreover, SecAlign's protection generalizes to strong attacks unseen in training. Specifically, the success rate of state-of-the-art GCG-based prompt injections drops from 56% to 2% in Mistral-7B after our alignment process. Our code is released at https://github.com/facebookresearch/SecAlign, Comment: Key words: prompt injection defense, LLM security, LLM-integrated applications. Alignment training makes LLMs robust against even the strongest prompt injection attacks

Radiopharmaceutical therapies (RPTs) present a major opportunity to improve cancer therapy. Although many current RPTs use the same injected activity for all patients, there is interest in using absorbed dose measurements to enable personalized prescriptions. Image-based absorbed dose calculations incur uncertainties from calibration factors, partial volume effects and segmentation methods. While previously published dose estimation protocols incorporate these uncertainties, they do not account for uncertainty originating from the reconstruction process itself with the propagation of Poisson noise from projection data. This effect should be accounted for to adequately estimate the total uncertainty in absorbed dose estimates. This paper proposes a computationally practical algorithm that propagates uncertainty from projection data through clinical reconstruction algorithms to obtain uncertainties on the total measured counts within volumes of interest (VOIs). The algorithm is first validated on ${}^{177}$Lu and ${}^{225}$Ac phantom data by comparing estimated uncertainties from individual SPECT acquisitions to empirical estimates obtained from multiple acquisitions. It is then applied to (i) Monte Carlo and (ii) multi-time point ${}^{177}$Lu-DOTATATE and ${}^{225}$Ac-PSMA-617 patient data for time integrated activity (TIA) uncertainty estimation. The outcomes of this work are two-fold: (i) the proposed uncertainty algorithm is validated, and (ii) a blueprint is established for how the algorithm can be inform dosimetry protocols via TIA uncertainty estimation. The proposed algorithm is made publicly available in the open-source image reconstruction library PyTomography.

Large Language Models (LLMs) are commonly used to generate solutions for mathematical reasoning problems in the following formats: natural language, code, or a combination of both. In this paper, we explore fundamental questions related to solving mathematical reasoning problems using natural language and code with state-of-the-art LLMs, including GPT-4o-mini and LLama-3.1-8b-Turbo. Our findings show that LLMs are better at reasoning in natural language compared to code. Additionally, although natural language and code serve as complementary forms of reasoning, they can affect each other in a negative way in certain scenarios. These insights motivate our development of a new prompting method, INC-Math, which leverages an LLM to dynamically select the most appropriate reasoning form, resulting in improved performance over comparable baselines with GPT-4o-mini.

In this work, we explore the decoding of mental imagery from subjects using their fMRI measurements. In order to achieve this decoding, we first created a mapping between a subject's fMRI signals elicited by the videos the subjects watched. This mapping associates the high dimensional fMRI activation states with visual imagery. Next, we prompted the subjects textually, primarily with emotion labels which had no direct reference to visual objects. Then to decode visual imagery that may have been in a person's mind's eye, we align a latent representation of these fMRI measurements with a corresponding video-fMRI based on textual labels given to the videos themselves. This alignment has the effect of overlapping the video fMRI embedding with the text-prompted fMRI embedding, thus allowing us to use our fMRI-to-video mapping to decode. Additionally, we enhance an existing fMRI dataset, initially consisting of data from five subjects, by including recordings from three more subjects gathered by our team. We demonstrate the efficacy of our model on this augmented dataset both in accurately creating a mapping, as well as in plausibly decoding mental imagery.

Neuroscience employs diverse neuroimaging techniques, each offering distinct insights into brain activity, from electrophysiological recordings such as EEG, which have high temporal resolution, to hemodynamic modalities such as fMRI, which have increased spatial precision. However, integrating these heterogeneous data sources remains a challenge, which limits a comprehensive understanding of brain function. We present the Spatiotemporal Alignment of Multimodal Brain Activity (SAMBA) framework, which bridges the spatial and temporal resolution gaps across modalities by learning a unified latent space free of modality-specific biases. SAMBA introduces a novel attention-based wavelet decomposition for spectral filtering of electrophysiological recordings, graph attention networks to model functional connectivity between functional brain units, and recurrent layers to capture temporal autocorrelations in brain signal. We show that the training of SAMBA, aside from achieving translation, also learns a rich representation of brain information processing. We showcase this classify external stimuli driving brain activity from the representation learned in hidden layers of SAMBA, paving the way for broad downstream applications in neuroscience research and clinical contexts.

IoT is rapidly growing from small-scale apps to large-scale apps. Small-scale apps employ short-range radios like Zigbee,BLE while large-scale apps employ long-range radios like LoRa,NB-IoT. The other upcoming category of apps like P2P energy-trade in smart homes are termed mesoscale IoT apps. There are no specialized radios for these apps. They either use short/long-range radios. To close this gap, we explored mesoscale apps using the COTS IoT radios available. Our qualitative analysis identifies Zigbee and LoRa as potential candidates. Our quantitative analysis on single and multi-hop topologies showed that Zigbee and LoRa achieve competitive throughput at a distance of 500-1200m from the gateway. A fundamental finding of these analyses is that a multi-radio system that can efficiently switch between Zigbee and LoRa performs better than the single-radio systems. However, instantaneously selecting and switching to a high-throughput radio during transmission is not trivial because of the erratic link quality dynamics. To address this issue, we developed MARS, that uses path quality metrics to instantaneously select the high-throughput radio during transmission. However, realizing MARS on resource-constrained end devices entails the challenge of obtaining instantaneous path-quality metrics. Traditional path quality estimation is not instantaneous due to propagation and queuing delays. We overcome this challenge by showing that collecting local path metrics as input to our decision trees provides sufficient information to instantaneously identify the high-throughput radio. The radio selector of MARS is powered by TAO-CART trees. The evaluation of MARS on a large-scale mesh topology at two different locations shows that MARS can efficiently identify and switch to the high-throughput radio during transmission, leading to an average throughput gain of 48.2% and 49.79% over its competitors.

Retrieval Augmented Generation (RAG) is a common method for integrating external knowledge into pretrained Large Language Models (LLMs) to enhance accuracy and relevancy in question answering (QA) tasks. However, prompt engineering and resource efficiency remain significant bottlenecks in developing optimal and robust RAG solutions for real-world QA applications. Recent studies have shown success in using fine tuning to address these problems; in particular, Retrieval Augmented Fine Tuning (RAFT) applied to smaller 7B models has demonstrated superior performance compared to RAG setups with much larger models such as GPT-3.5. The combination of RAFT with parameter-efficient fine tuning (PEFT) techniques, such as Low-Rank Adaptation (LoRA), promises an even more efficient solution, yet remains an unexplored area. In this work, we combine RAFT with LoRA to reduce fine tuning and storage requirements and gain faster inference times while maintaining comparable RAG performance. This results in a more compute-efficient RAFT, or CRAFT, which is particularly useful for knowledge-intensive QA tasks in resource-constrained environments where internet access may be restricted and hardware resources limited., Comment: 6 pages, 2 tables

This paper introduces a Transformer-Enhanced Physics-Informed Neural Network (TE-PINN) designed for accurate quaternion-based orientation estimation in high-dynamic environments, particularly within the field of robotics. By integrating transformer networks with physics-informed learning, our approach innovatively captures temporal dependencies in sensor data while enforcing the fundamental physical laws governing rotational motion. TE-PINN leverages a multi-head attention mechanism to handle sequential data from inertial sensors, such as accelerometers and gyroscopes, ensuring temporal consistency. Simultaneously, the model embeds quaternion kinematics and rigid body dynamics into the learning process, aligning the network's predictions with mechanical principles like Euler's laws of motion. The physics-informed loss function incorporates the dynamics of angular velocity and external forces, enhancing the network's ability to generalize in complex scenarios. Our experimental evaluation demonstrates that TE-PINN consistently outperforms traditional methods such as Extended Kalman Filters (EKF) and LSTM-based estimators, particularly in scenarios characterized by high angular velocities and noisy sensor data. The results show a significant reduction in mean quaternion error and improved gyroscope bias estimation compared to the state-of-the-art. An ablation study further isolates the contributions of both the transformer architecture and the physics-informed constraints, highlighting the synergistic effect of both components in improving model performance. The proposed model achieves real-time performance on embedded systems typical of mobile robots, offering a scalable and efficient solution for orientation estimation in autonomous systems.

We demonstrate that the Wasserstein distance $W_{1}$ corresponding to optical tomograms of nonclassical states faithfully captures changes that arise due to photon addition to, or subtraction from, these states. $W_{1}$ is a true measure of distance in the quantum state space, and is sensitive to the underlying interference structures that arise in the tomogram after changes in the photon number. Our procedure is universally applicable to the cat and squeezed states, the former displaying the characteristic negativity in its Wigner function, while the latter does not do so. We explicate this in the case of the squeezed vacuum and even coherent states and show that photon addition (or subtraction) is mirrored in the shift in the intensity of specific regions in the tomogram. Further, we examine the dependence of $W_{1}$ on the squeezing parameter, and its sensitivity to different quadratures., Comment: 10 pages, 12 figures

We demonstrate that (001) grown Cd3As2 thin films with a superlattice-patterned gate can potentially realize the moir\'e Bernevig-Hughes-Zhang (BHZ) model. Our calculations identify the parameterization region necessary to achieve topological flat mini-bands with a C4z symmetric and a C6z symmetric potential. Additionally, we show that a spin-polarized state can serve as the minimal platform for hosting the moir\'e induced quantum anomalous Hall effect, supported by Hartree Fock interaction kernel analysis and self-consistent mean field calculations., Comment: 10 pages, 10 figures

Aerial manipulation combines the versatility and speed of flying platforms with the functional capabilities of mobile manipulation, which presents significant challenges due to the need for precise localization and control. Traditionally, researchers have relied on offboard perception systems, which are limited to expensive and impractical specially equipped indoor environments. In this work, we introduce a novel platform for autonomous aerial manipulation that exclusively utilizes onboard perception systems. Our platform can perform aerial manipulation in various indoor and outdoor environments without depending on external perception systems. Our experimental results demonstrate the platform's ability to autonomously grasp various objects in diverse settings. This advancement significantly improves the scalability and practicality of aerial manipulation applications by eliminating the need for costly tracking solutions. To accelerate future research, we open source our ROS 2 software stack and custom hardware design, making our contributions accessible to the broader research community., Comment: Project website: https://sites.google.com/view/open-source-soft-platform/open-source-soft-robotic-platform

Modeling of the collimator-detector response (CDR) in SPECT reconstruction enables improved resolution and more accurate quantitation, especially for higher energy imaging (e.g.Lu-177 and Ac-225). Such modeling, however, can pose a significant computational bottleneck when there are substantial components of septal penetration and scatter in the acquired data, since a direct convolution-based approach requires large 2D kernels. The present work presents an alternative method for fast and accurate CDR compensation using a linear operator built from 1D convolutions and rotations (1D-R). To enable open-source development and use of these models in image reconstruction, we release a SPECTPSFToolbox repository for the PyTomography project on GitHub.

Information retrieval methods often rely on a single embedding model trained on large, general-domain datasets like MSMARCO. While this approach can produce a retriever with reasonable overall performance, models trained on domain-specific data often yield better results within their respective domains. While prior work in information retrieval has tackled this through multi-task training, the topic of combining multiple domain-specific expert retrievers remains unexplored, despite its popularity in language model generation. In this work, we introduce RouterRetriever, a retrieval model that leverages multiple domain-specific experts along with a routing mechanism to select the most appropriate expert for each query. It is lightweight and allows easy addition or removal of experts without additional training. Evaluation on the BEIR benchmark demonstrates that RouterRetriever outperforms both MSMARCO-trained (+2.1 absolute nDCG@10) and multi-task trained (+3.2) models. This is achieved by employing our routing mechanism, which surpasses other routing techniques (+1.8 on average) commonly used in language modeling. Furthermore, the benefit generalizes well to other datasets, even in the absence of a specific expert on the dataset. To our knowledge, RouterRetriever is the first work to demonstrate the advantages of using multiple domain-specific expert embedding models with effective routing over a single, general-purpose embedding model in retrieval tasks.

We make use of model-independent statistical methods to assess the consistency of three different supernova compilations: Union3, Pantheon+ and DES 5-year supernovae. We expand the available model space of each, using Crossing Statistics, and test the compatibility of each dataset, against the other two. This is done using (I) a Flat $\Lambda$CDM fitting to, and (II) Iterative Smoothing from, one particular dataset, and determining the level of deformation by required to fit the other two. This allows us to test the mutual consistency of the datasets both within the standard model and in the case of some extended model, motivated by features present in a particular dataset. We find that, in both these cases, the data are only consistent with the point in the parameter space corresponding to zero deformation, at around a $2\sigma$ level, with the DES compilation showing the largest disagreement. However, all three datasets are still found to be consistent to within $1-2\sigma$ for some subset of the extended model space implied by the deformations., Comment: 36 pages, 15 figures

Quasars could serve as standard candles if the relation between their ultraviolet and X-ray luminosities can be accurately calibrated. Previously, we developed a model-independent method to calibrate quasar standard candles using the distances-redshift relation reconstructed from Type Ia supernova at z<2 using Gaussian process regression. Interestingly, we found that the calibrated quasar standard candle dataset preferred a deviation from $\Lambda$CDM at redshifts above z>2. One interpretation of these findings is that the calibration parameters of the quasar UV-X-ray luminosity relationship evolves with redshift. In order to test the redshift dependence of the quasar calibration in a model-independent manner, we divided the quasar sample whose redshift overlap with the redshift coverage of Pantheon+ Type Ia supernova compilation into two sub-samples: a low-redshift quasar sub-sample and a high-redshift quasar sub-sample. Our present results show that there is about a 4$\sigma$ inconsistency between the quasar parameters inferred from the high-redshift quasar sub-sample and from the low-redshift sub-sample if no evolution of the quasar relation is considered. This inconsistency suggests the necessity of considering redshift evolution for the relationship between the quasars$'$ ultraviolet and X-ray luminosities. We then test an explicit parametrization of the redshift evolution of the quasar calibration parameters via $\gamma(z) = \gamma_0+\gamma_1(1+z)$ and $\beta(z)=\beta_0+\beta_1(1+z)$. Combining this redshift-dependent calibration relationship with the distance-redshift relationship reconstructed from Pantheon+ supernova compilation, we find the high-redshift sub-sample and low-redshift sub-sample become consistent at the 1$\sigma$ level, which means that the parameterized form of $\gamma(z)$ and $\beta(z)$ works well at describing the evolution of the quasar calibration parameters., Comment: 8 pages, 3 figures

Let $S_l(M,N)$ denote a set of $\ell$ triples of positive integers whose parts have the same sum $M$ and the same product $N$. For each $2\leq\ell\leq 4$ we establish a connection between the set $S_l(M,N)$ and a family of elliptic curves. When $\ell=2$ and $3$, we use certain known parametrised sets $S_l(M,N)$ and respectively find families of elliptic curves of generic rank~$\geq 5$ and $\geq 6$, while for $\ell=4$ we first obtain a parametrised set $S_l(M,N)$, then construct a family of elliptic curves of generic rank $\geq 8$. Finally, we perform a computer search within these families to find specific curves with rank~$\geq 11$. The highest rank examples we found were two curves of rank~$14$., Comment: 11 pages, 4 tables

Polyurea (PUa) elastomers are extensively used in a wide range of applications spanning from biomedical to defense fields due to their enabling mechanical properties. These materials can be further reinforced through the incorporation of nanoparticles to form nanocomposites. This study focuses on an IPDI-based PUa matrix with exfoliated graphene nanoplatelet (xGnP) fillers. We propose a generalized constitutive model by integrating one Fractional Maxwell Model (FMM) and one Fractional Maxwell Gel (FMG) branch in a parallel configuration via introducing a new dimensionless number to bridge between these branches physically and mathematically. Through systematic local-to-global sensitivity analyses, we investigate the behavior of these nanocomposites to facilitate simulation, design, and performance prediction. Consistently, the constructed models share the same most/least influential model parameters. $\alpha_1$ and $E_{c_1}$, the power exponent and the characteristic modulus of the first branch, are found to be the most influential model parameters, while $\tau_{c_2}$ and $\tau_{c_1}$, the characteristic time-scales of each branch, are recognized as the least influential model parameters. The proposed PU nanocomposite constitutive laws can now make an impact to the design and optimization of coating and shock-absorbing coatings in a range of applications.

Sub-picosecond timing synchronization can enable future optical timekeeping networks, including coherent phased array radar imaging at GHz levels, intercontinental clock comparisons for the redefinition of the second, chronometric leveling, and synchronization of remote assets, including future satellite-based optical time standards. With optical clocks now operating on mobile platforms, free-space synchronization networks with compatible performance and the ability to operate under platform motion are essential to expand the reach of precision timing. Recently, femtosecond (fs)-level optical time-transfer techniques have been developed that can operate over hundreds of kilometers despite atmospheric turbulence, signal fade, and dropouts. Here we report a two-tone optical time transfer scheme with comparable performance that reduces hardware requirements and can support both fiber and free-space networks. Using this technique, sub-fs synchronization was demonstrated over a $\sim$100 m free-space link for several hours. In addition, the link was used to syntonize two iodine optical clocks and then compare them over four days. The set-up employs an integrated photonics transceiver and telecom-band lasers that are compatible with full photonic integration.

In this paper, we present a novel cryptographic system that integrates Quantum Key Distribution (QKD) with classical encryption techniques to secure steganographic images. Our approach leverages the E91 QKD protocol to generate a shared secret key between communicating parties, ensuring the highest level of security against eavesdropping through the principles of quantum mechanics. This key is then hashed using the Secure Hash Algorithm (SHA) to provide a fixedlength, high-entropy key, which is subsequently utilized in symmetric encryption. We explore the use of AES (Advanced Encryption Standard) algorithms for encrypting steganographic images, which hide sensitive information within digital images to provide an additional layer of security through obscurity. The combination of QKD, hashing, and symmetric encryption offers a robust security framework that mitigates various attack vectors, enhancing the confidentiality and integrity of the transmitted data. Our experimental results demonstrate the feasibility and efficiency of the proposed system, highlighting its performance in terms of key generation rates, encryption/decryption speeds, and the computational overhead introduced by the hashing and steganographic processes. By integrating quantum and classical cryptographic methods with steganography, this work provides a comprehensive security solution that is highly resistant to both quantum and classical attacks, making it suitable for applications requiring stringent security measures. This paper contributes to the ongoing research in cryptographic systems, offering insights into the practical implementation and potential benefits of hybrid quantumclassical security protocols., Comment: 18 pages, 13 figures, 46 equations

In this paper we extend some set theoretic concepts of numerical semigroups for arbitrary sub-semigroups of natural numbers. Then we characterized gapsets which leads to a more efficient computational approach towards numerical semigroups and finally we introduce the extension of gapsets and prove that the sequence of the number of gapsets of size $g$ is non-decreasing as a weak version of Bras-Amor\'os's conjecture., Comment: 14 pages, 2 algorithms

We show that any finite family of pairwise intersecting balls in $\mathbb{E}^n$ can be pierced by $(\sqrt{3/2}+o(1))^n$ points improving the previously known estimate of $(2+o(1))^n$. As a corollary, this implies that any $2$-illuminable spiky ball in $\mathbb{E}^n$ can be illuminated by $(\sqrt{3/2}+o(1))^n$ directions. For the illumination number of convex spiky balls, i.e., cap bodies, we show an upper bound in terms of the sizes of certain related spherical codes and coverings. For large dimensions, this results in an upper bound of $1.19851^n$, which can be compared with the previous $(\sqrt{2}+o(1))^n$ established only for the centrally symmetric cap bodies. We also prove the lower bounds of $(\tfrac{2}{\sqrt{3}}-o(1))^n$ for the three problems above.

Vertebrate animals benefit from a combination of rigidity for structural support and softness for adaptation. Similarly, integrating rigidity and softness can enhance the versatility of soft robotics. However, the challenges associated with creating durable bonding interfaces between soft and rigid materials have limited the development of hybrid robots. Existing solutions require specialized machinery, such as polyjet 3D printers, which are not commonly available. In response to these challenges, we have developed a 3D printing technique that can be used with almost all commercially available FDM printers. This technique leverages the common issue of underextrusion to create a strong bond between soft and rigid materials. Underextrusion generates a porous structure, similar to fibrous connective tissues, that provides a robust interface with the rigid part through layer fusion, while the porosity enables interlocking with the soft material. Our experiments demonstrated that this method outperforms conventional adhesives commonly used in soft robotics, achieving nearly 200\% of the bonding strength in both lap shear and peeling tests. Additionally, we investigated how different porosity levels affect bonding strength. We tested the technique under pressure scenarios critical to soft and hybrid robots and achieved three times more pressure than the current adhesion solution. Finally, we fabricated various hybrid robots using this technique to demonstrate the wide range of capabilities this approach and hybridity can bring to soft robotics. has context menu

Purpose: Artificial intelligence (AI) techniques have been extensively utilized for diagnosing and prognosis of several diseases in recent years. This study identifies, appraises and synthesizes published studies on the use of AI for the prognosis of COVID-19. Method: Electronic search was performed using Medline, Google Scholar, Scopus, Embase, Cochrane and ProQuest. Studies that examined machine learning or deep learning methods to determine the prognosis of COVID-19 using CT or chest X-ray images were included. Polled sensitivity, specificity area under the curve and diagnostic odds ratio were calculated. Result: A total of 36 articles were included; various prognosis-related issues, including disease severity, mechanical ventilation or admission to the intensive care unit and mortality, were investigated. Several AI models and architectures were employed, such as the Siamense model, support vector machine, Random Forest , eXtreme Gradient Boosting, and convolutional neural networks. The models achieved 71%, 88% and 67% sensitivity for mortality, severity assessment and need for ventilation, respectively. The specificity of 69%, 89% and 89% were reported for the aforementioned variables. Conclusion: Based on the included articles, machine learning and deep learning methods used for the prognosis of COVID-19 patients using radiomic features from CT or CXR images can help clinicians manage patients and allocate resources more effectively. These studies also demonstrate that combining patient demographic, clinical data, laboratory tests and radiomic features improves model performances.

Recent research endeavours have theoretically shown the beneficial effect of cooperation in multi-agent reinforcement learning (MARL). In a setting involving $N$ agents, this beneficial effect usually comes in the form of an $N$-fold linear convergence speedup, i.e., a reduction - proportional to $N$ - in the number of iterations required to reach a certain convergence precision. In this paper, we show for the first time that this speedup property also holds for a MARL framework subject to asynchronous delays in the local agents' updates. In particular, we consider a policy evaluation problem in which multiple agents cooperate to evaluate a common policy by communicating with a central aggregator. In this setting, we study the finite-time convergence of \texttt{AsyncMATD}, an asynchronous multi-agent temporal difference (TD) learning algorithm in which agents' local TD update directions are subject to asynchronous bounded delays. Our main contribution is providing a finite-time analysis of \texttt{AsyncMATD}, for which we establish a linear convergence speedup while highlighting the effect of time-varying asynchronous delays on the resulting convergence rate.

The ability to form diamond electrodes on insulating polycrystalline diamond substrates using single-step laser patterning, and the use of the electrodes as a substrate that supports the adhesion and proliferation of human mesenchymal stem cells (hMSCs) is demonstrated. Laser induced graphitisation results in a conductive amorphous carbon surface, rich in oxygen and nitrogen terminations. This results in an electrode with a high specific capacitance of 182 uF/cm2, a wide water window of 3.25 V, and a low electrochemical impedance of 129 Ohms/cm2 at 1 kHz. The electrodes surface exhibited a good level of biocompatibility with hMSCs, supporting cell adhesion and proliferation. The cells cultured on the electrode displayed significant elongation and alignment along the direction of the laser patterned microgrooves. Because of its favourable electrochemical performance and biocompatibility, the laser-patterned diamond electrodes create a potential for a versatile platform in stem cell therapeutics.

Prostate specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT) imaging provides a tremendously exciting frontier in visualization of prostate cancer (PCa) metastatic lesions. However, accurate segmentation of metastatic lesions is challenging due to low signal-to-noise ratios and variable sizes, shapes, and locations of the lesions. This study proposes a novel approach for automated segmentation of metastatic lesions in PSMA PET/CT 3D volumetric images using 2D denoising diffusion probabilistic models (DDPMs). Instead of 2D trans-axial slices or 3D volumes, the proposed approach segments the lesions on generated multi-angle maximum intensity projections (MA-MIPs) of the PSMA PET images, then obtains the final 3D segmentation masks from 3D ordered subset expectation maximization (OSEM) reconstruction of 2D MA-MIPs segmentations. Our proposed method achieved superior performance compared to state-of-the-art 3D segmentation approaches in terms of accuracy and robustness in detecting and segmenting small metastatic PCa lesions. The proposed method has significant potential as a tool for quantitative analysis of metastatic burden in PCa patients., Comment: 11 pages, 2 figures, accepted in the DGM4MICCAI workshop, MICCAI, 2024

With the rise of tiny IoT devices powered by machine learning (ML), many researchers have directed their focus toward compressing models to fit on tiny edge devices. Recent works have achieved remarkable success in compressing ML models for object detection and image classification on microcontrollers with small memory, e.g., 512kB SRAM. However, there remain many challenges prohibiting the deployment of ML systems that require high-resolution images. Due to fundamental limits in memory capacity for tiny IoT devices, it may be physically impossible to store large images without external hardware. To this end, we propose a high-resolution image scaling system for edge ML, called HiRISE, which is equipped with selective region-of-interest (ROI) capability leveraging analog in-sensor image scaling. Our methodology not only significantly reduces the peak memory requirements, but also achieves up to 17.7x reduction in data transfer and energy consumption., Comment: 10 pages, 8 figures

Direct Preference Optimization (DPO) has become a widely used training method for the instruction fine-tuning of large language models (LLMs). In this work, we explore an under-investigated aspect of DPO - its dependency on the reference model or policy. Such reference policies, typically instantiated as the model to be further fine-tuned, are important since they can impose an upper limit on DPO's effectiveness. Therefore, we address three related research questions in this work. First, we explore the optimal strength of the KL divergence constraint in DPO, which penalizes deviations from the reference policy, and find that DPO is sensitive to this strength. Next, we examine the necessity of the KL-constraint from the reference policies in DPO by providing both theoretical and empirical comparisons between DPO and related learning objectives, demonstrating DPO's superiority in this controlled setting. Additionally, we investigate whether DPO benefits from stronger reference policies, finding that a stronger reference policy can lead to improved performance, but only when it is similar to the model being fine-tuned. Our findings highlight the confounding role of reference policies in DPO and offer insights for best practices, while also identifying open research questions for future studies., Comment: GitHub Repo: https://github.com/yale-nlp/refdpo

The European Committee for Future Accelerators (ECFA) Early-Career Researcher (ECR) panel, which represents the interests of the ECR community to ECFA, presents in this document its initiatives and activities in the year 2023. This report summarises the process of the first big turnover in the panel composition at the start of 2023 and reports on the activities of the active working groups - either pursued from before or newly established. The overarching goal of the ECFA-ECR panel is to better understand and support the diverse interests of early-career researchers in the ECFA community and beyond., Comment: Editors: Jan-Hendrik Arling, Cecilia Borca, Armin Ilg, Arnau Morancho Tarda, Holly Pacey, Marko Pesut, Elisabetta Spadaro Norella and Marta Urbaniak. arXiv admin note: substantial text overlap with arXiv:2212.11238

Matroid theory is fundamentally connected with index coding and network coding problems. In fact, the reliance of linear index coding and network coding rates on the characteristic of a field has been demonstrated by using the two well-known matroid instances, namely the Fano and non-Fano matroids. This established the insufficiency of linear coding, one of the fundamental theorems in both index coding and network coding. While the Fano matroid is linearly representable only over fields with characteristic two, the non-Fano instance is linearly representable only over fields with odd characteristic. For fields with arbitrary characteristic $p$, the Fano and non-Fano matroids were extended to new classes of matroid instances whose linear representations are dependent on fields with characteristic $p$. However, these matroids have not been well appreciated nor cited in the fields of network coding and index coding. In this paper, we first reintroduce these matroids in a more structured way. Then, we provide a completely independent alternative proof with the main advantage of using only matrix manipulation rather than complex concepts in number theory and matroid theory. In this new proof, it is shown that while the class $p$-Fano matroid instances are linearly representable only over fields with characteristic $p$, the class $p$-non-Fano instances are representable over fields with any characteristic other than characteristic $p$. Finally, following the properties of the class $p$-Fano and $p$-non-Fano matroid instances, we characterize two new classes of index coding instances, respectively, referred to as the class $p$-Fano and $p$-non-Fano index coding, each with a size of $p^2 + 4p + 3$., Comment: arXiv admin note: text overlap with arXiv:2201.10057

The objective of this study was to develop an automated pipeline that enhances thyroid disease classification using thyroid scintigraphy images, aiming to decrease assessment time and increase diagnostic accuracy. Anterior thyroid scintigraphy images from 2,643 patients were collected and categorized into diffuse goiter (DG), multinodal goiter (MNG), and thyroiditis (TH) based on clinical reports, and then segmented by an expert. A ResUNet model was trained to perform auto-segmentation. Radiomic features were extracted from both physician (scenario 1) and ResUNet segmentations (scenario 2), followed by omitting highly correlated features using Spearman's correlation, and feature selection using Recursive Feature Elimination (RFE) with XGBoost as the core. All models were trained under leave-one-center-out cross-validation (LOCOCV) scheme, where nine instances of algorithms were iteratively trained and validated on data from eight centers and tested on the ninth for both scenarios separately. Segmentation performance was assessed using the Dice similarity coefficient (DSC), while classification performance was assessed using metrics, such as precision, recall, F1-score, accuracy, area under the Receiver Operating Characteristic (ROC AUC), and area under the precision-recall curve (PRC AUC). ResUNet achieved DSC values of 0.84$\pm$0.03, 0.71$\pm$0.06, and 0.86$\pm$0.02 for MNG, TH, and DG, respectively. Classification in scenario 1 achieved an accuracy of 0.76$\pm$0.04 and a ROC AUC of 0.92$\pm$0.02 while in scenario 2, classification yielded an accuracy of 0.74$\pm$0.05 and a ROC AUC of 0.90$\pm$0.02. The automated pipeline demonstrated comparable performance to physician segmentations on several classification metrics across different classes, effectively reducing assessment time while maintaining high diagnostic accuracy. Code available at: https://github.com/ahxmeds/thyroidiomics.git., Comment: 7 pages, 4 figures, 2 tables

Graph neural network (GNN) explainers identify the important subgraph that ensures the prediction for a given graph. Until now, almost all GNN explainers are based on association, which is prone to spurious correlations. We propose {\name}, a GNN causal explainer via causal inference. Our explainer is based on the observation that a graph often consists of a causal underlying subgraph. {\name} includes three main steps: 1) It builds causal structure and the corresponding structural causal model (SCM) for a graph, which enables the cause-effect calculation among nodes. 2) Directly calculating the cause-effect in real-world graphs is computationally challenging. It is then enlightened by the recent neural causal model (NCM), a special type of SCM that is trainable, and design customized NCMs for GNNs. By training these GNN NCMs, the cause-effect can be easily calculated. 3) It uncovers the subgraph that causally explains the GNN predictions via the optimized GNN-NCMs. Evaluation results on multiple synthetic and real-world graphs validate that {\name} significantly outperforms existing GNN explainers in exact groundtruth explanation identification

In this analysis we apply a model-independent framework to test the flat $\Lambda$CDM cosmology using simulated SNIa data from the upcoming Legacy Survey of Space and Time (LSST) and combined with simulated Dark Energy Spectroscopic Instrument (DESI) five-years Baryon Acoustic Oscillations (BAO) data. We adopt an iterative smoothing technique to reconstruct the expansion history from SNIa data, which, when combined with BAO measurements, facilitates a comprehensive test of the Universe's curvature and the nature of dark energy. The analysis is conducted under three different mock true cosmologies: a flat $\Lambda$CDM universe, a universe with a notable curvature ($\Omega_{k,0} = 0.1$), and one with dynamically evolving dark energy. Each cosmology demonstrates different kinds and varying degrees of deviation from the standard model predictions. We forecast that our reconstruction technique can constrain cosmological parameters, such as the curvature ($\Omega_{k,0}$) and $c/H_0 r_\mathrm{d}$, with a precision of approximately 0.5\% for $c/H_0r_\mathrm{d}$ and 0.04 for $\Omega_{k,0}$, competitive with current cosmic microwave background constraints, without assuming any form of dark energy.