Search

Measurements of the EMC effect in the tritium and helium-3 mirror nuclei are reported. The data were obtained by the MARATHON Jefferson Lab experiment, which performed deep inelastic electron scattering from deuterium and the three-body nuclei, using a cryogenic gas target system and the High Resolution Spectrometers of the Hall A Facility of the Lab. The data cover the Bjorken $x$ range from 0.20 to 0.83, corresponding to a squared four-momentum transfer $Q^2$ range from 2.7 to $11.9\gevsq$, and to an invariant mass $W$ of the final hadronic state greater than 1.84 GeV/${\it c}^2$. The tritium EMC effect measurement is the first of its kind. The MARATHON experimental results are compared to results from previous measurements by DESY-HERMES and JLab-Hall C experiments, as well as with few-body theoretical predictions., Comment: arXiv admin note: text overlap with arXiv:2104.05850

A measurement of the double-spin asymmetry from electron-$^{3}$He scattering in the threshold region of two- and three-body breakup of $^{3}$He was performed at Jefferson Lab, for Q$^{2}$ values of 0.1 and 0.2 (GeV/$c$)$^{2}$. The results of this measurement serve as a stringent test of our understanding of few-body systems. When compared with calculations from plane wave impulse approximation and Faddeev theory, we found that the Faddeev calculations, which use modern nuclear potentials and prescriptions for meson-exchange currents, demonstrate an overall good agreement with data.

Entanglement fluctuations associated with Schr\"{o}dinger evolution of wavefunctions offer a unique perspective on various fundamental issues ranging from quantum thermalization to state preparation in quantum devices. Very recently, a subset of present authors have shown that in a class of free-fermion lattice models and interacting spin chains, entanglement dynamics enters into a new regime at long time, with entanglement probes displaying persistent temporal fluctuations, whose statistics falls into the seemingly disparate paradigm of mesoscopic fluctuations in condensed matter physics. This motivate us to revisit here entanglement dynamics of a canonical bosonic model in many-body physics, i.e., a coupled harmonic oscillator chain. We find that when the system is driven out of equilibrium, the long-time entanglement dynamics exhibits strictly the same statistical behaviors as that of free-fermion models. Specifically, irrespective of entanglement probes and microscopic parameters, the statistical distribution of entanglement fluctuations is flanked by asymmetric tails: sub-Gaussian for upward fluctuations and sub-Gamma for downward; moreover, the variance exhibits a crossover from the scaling $\sim 1/L$ to $\sim L_A^3/L^2$, as the subsystem size $L_A$ increases ($L$ the total system size). This insensitivity to the particle statistics, dubbed boson-fermion universality, is contrary to the common wisdom that statistical phenomena of many-body nature depend strongly on particle statistics. Together with our previous work, the present work indicates rich fluctuation phenomena in entanglement dynamics awaiting in-depth explorations., Comment: 18 pages, 7 figures

Ferroelectric tunnel junctions (FTJs) harness the unique combination of ferroelectricity and quantum tunneling, and thus herald new opportunities in next-generation nonvolatile memory technologies. Recent advancements in the fabrication of ultrathin heterostructures have enabled the integration of ferroelectrics with various functional materials, forming hybrid tunneling-diode junctions. These junctions benefit from the modulation of the functional layer/ferroelectric interface through ferroelectric polarization, thus enabling further modalities and functional capabilities than in addition to tunneling electroresistance. This perspective aims to provide in-depth insight into novel physical phenomena of several typical ferroelectric hybrid junctions, ranging from ferroelectric/dielectric, ferroelectric/multiferroic, ferroelectric/superconducting to ferroelectric/2D materials, and finally their expansion into the realm of ferroelectric resonant tunneling diodes (FeRTDs). This latter aspect, i.e., resonant tunneling offers a radically new approach to exploiting tunneling behavior in ferroelectric heterostructures. We discuss examples that have successfully shown room temperature ferroelectric control of parameters such as the resonant peak, tunnel current ratio at peak and negative differential resistance. We conclude the perspective by summarizing the challenges and highlighting the opportunities for the future development of hybrid FTJs with a special emphasis on a new possible type of FeRTD device. The prospects for enhanced performance and expanded functionality ignite tremendous excitement in hybrid FTJs and FeRTDs for future nanoelectronics.

On May 24th, 2023, the Advanced Laser Interferometer Gravitational-Wave Observatory (LIGO), joined by the Advanced Virgo and KAGRA detectors, began the fourth observing run for a two-year-long dedicated search for gravitational waves. The LIGO Hanford and Livingston detectors have achieved an unprecedented sensitivity to gravitational waves, with an angle-averaged median range to binary neutron star mergers of 152 Mpc and 160 Mpc, and duty cycles of 65.0% and 71.2%, respectively, with a coincident duty cycle of 52.6%. The maximum range achieved by the LIGO Hanford detector is 165 Mpc and the LIGO Livingston detector 177 Mpc, both achieved during the second part of the fourth observing run. For the fourth run, the quantum-limited sensitivity of the detectors was increased significantly due to the higher intracavity power from laser system upgrades and replacement of core optics, and from the addition of a 300 m filter cavity to provide the squeezed light with a frequency-dependent squeezing angle, part of the A+ upgrade program. Altogether, the A+ upgrades led to reduced detector-wide losses for the squeezed vacuum states of light which, alongside the filter cavity, enabled broadband quantum noise reduction of up to 5.2 dB at the Hanford observatory and 6.1 dB at the Livingston observatory. Improvements to sensors and actuators as well as significant controls commissioning increased low frequency sensitivity. This paper details these instrumental upgrades, analyzes the noise sources that limit detector sensitivity, and describes the commissioning challenges of the fourth observing run., Comment: 26 pages, 18 figures

Light echoes offer a means of studying protoplanetary disks, including their geometry and composition, even when they are not spatially resolved. We present a test of this approach applied specifically to optically thick, geometrically flared disks around active stars. Here we adopt stellar parameters of an active M dwarf to calculate light echoes for disks and rings with radii that would produce time delays consistent with TESS short cadence (about 2 minutes) time bins. Our results show successful fits to disk parameters, highlighting the potential effectiveness of this method in the search for protoplanetary disks., Comment: 7 pages, 3 figures

Transitioning from quantum computation on physical qubits to quantum computation on encoded, logical qubits can improve the error rate of operations, and will be essential for realizing valuable quantum computational advantages. Using a neutral atom quantum processor with 256 qubits, each an individual Ytterbium atom, we demonstrate the entanglement of 24 logical qubits using the distance-two [[4,2,2]] code, simultaneously detecting errors and correcting for lost qubits. We also implement the Bernstein-Vazirani algorithm with up to 28 logical qubits encoded in the [[4,1,2]] code, showing better-than-physical error rates. We demonstrate fault-tolerant quantum computation in our approach, guided by the proposal of Gottesman (2016), by performing repeated loss correction for both structured and random circuits encoded in the [[4,2,2]] code. Finally, since distance-two codes can correct qubit loss, but not other errors, we show repeated loss and error correction using the distance-three [[9,1,3]] Bacon-Shor code. These results begin to clear a path for achieving scientific quantum advantage with a programmable neutral atom quantum processor., Comment: 17 pages, 16 figures

Arrays of optically trapped neutral atoms are a promising architecture for the realization of quantum computers. In order to run increasingly complex algorithms, it is advantageous to demonstrate high-fidelity and flexible gates between long-lived and highly coherent qubit states. In this work, we demonstrate a universal high-fidelity gate-set with individually controlled and parallel application of single-qubit gates and two-qubit gates operating on the ground-state nuclear spin qubit in arrays of tweezer-trapped $^{171}$Yb atoms. We utilize the long lifetime, flexible control, and high physical fidelity of our system to characterize native gates using single and two-qubit Clifford and symmetric subspace randomized benchmarking circuits with more than 200 CZ gates applied to one or two pairs of atoms. We measure our two-qubit entangling gate fidelity to be 99.72(3)% (99.40(3)%) with (without) post-selection. In addition, we introduce a simple and optimized method for calibration of multi-parameter quantum gates. These results represent important milestones towards executing complex and general quantum computation with neutral atoms.

Splenomegaly, the enlargement of the spleen, is an important clinical indicator for various associated medical conditions, such as sickle cell disease (SCD). Spleen length measured from 2D ultrasound is the most widely used metric for characterising spleen size. However, it is still considered a surrogate measure, and spleen volume remains the gold standard for assessing spleen size. Accurate spleen volume measurement typically requires 3D imaging modalities, such as computed tomography or magnetic resonance imaging, but these are not widely available, especially in the Global South which has a high prevalence of SCD. In this work, we introduce a deep learning pipeline, DeepSPV, for precise spleen volume estimation from single or dual 2D ultrasound images. The pipeline involves a segmentation network and a variational autoencoder for learning low-dimensional representations from the estimated segmentations. We investigate three approaches for spleen volume estimation and our best model achieves 86.62%/92.5% mean relative volume accuracy (MRVA) under single-view/dual-view settings, surpassing the performance of human experts. In addition, the pipeline can provide confidence intervals for the volume estimates as well as offering benefits in terms of interpretability, which further support clinicians in decision-making when identifying splenomegaly. We evaluate the full pipeline using a highly realistic synthetic dataset generated by a diffusion model, achieving an overall MRVA of 83.0% from a single 2D ultrasound image. Our proposed DeepSPV is the first work to use deep learning to estimate 3D spleen volume from 2D ultrasound images and can be seamlessly integrated into the current clinical workflow for spleen assessment., Comment: arXiv admin note: substantial text overlap with arXiv:2308.08038

We developed a shoe-mounted gait monitoring system capable of tracking up to 17 gait parameters, including gait length, step time, stride velocity, and others. The system employs a stereo camera mounted on one shoe to track a marker placed on the opposite shoe, enabling the estimation of spatial gait parameters. Additionally, a Force Sensitive Resistor (FSR) affixed to the heel of the shoe, combined with a custom-designed algorithm, is utilized to measure temporal gait parameters. Through testing on multiple participants and comparison with the gait mat, the proposed gait monitoring system exhibited notable performance, with the accuracy of all measured gait parameters exceeding 93.61%. The system also demonstrated a low drift of 4.89% during long-distance walking. A gait identification task conducted on participants using a trained Transformer model achieved 95.7% accuracy on the dataset collected by the proposed system, demonstrating that our hardware has the potential to collect long-sequence gait data suitable for integration with current Large Language Models (LLMs). The system is cost-effective, user-friendly, and well-suited for real-life measurements., Comment: 13 pages, 14 figures. This paper was submitted for publication to the IEEE Transactions on Instrumentation and Measurement

We propose to measure the beam normal single spin asymmetry in elastic scattering of transversely polarized electron from target nuclei with 12 $\leq Z \leq$ 90 at Q$^2$ = 0.0092 GeV$^2$ to study its nuclear dependence. While the theoretical calculations based on two-photon exchange suggest no nuclear dependence at this kinematics, the results of 208Pb from Jefferson Lab show a striking disagreement from both theoretical predictions and light nuclei measurements. The proposed measurements will provide new data for intermediate to heavy nuclei where no data exists for $Z \geq$ 20 in the kinematics of previous high-energy experiments. It will allow one to investigate the missing contributions that are not accounted in the current theoretical models., Comment: Submitted to Jefferson Lab PAC52

The Knowledge Graph Completion~(KGC) task aims to infer the missing entity from an incomplete triple. Existing embedding-based methods rely solely on triples in the KG, which is vulnerable to specious relation patterns and long-tail entities. On the other hand, text-based methods struggle with the semantic gap between KG triples and natural language. Apart from triples, entity contexts (e.g., labels, descriptions, aliases) also play a significant role in augmenting KGs. To address these limitations, we propose KGR3, a context-enriched framework for KGC. KGR3 is composed of three modules. Firstly, the Retrieval module gathers supporting triples from the KG, collects plausible candidate answers from a base embedding model, and retrieves context for each related entity. Then, the Reasoning module employs a large language model to generate potential answers for each query triple. Finally, the Re-ranking module combines candidate answers from the two modules mentioned above, and fine-tunes an LLM to provide the best answer. Extensive experiments on widely used datasets demonstrate that KGR3 consistently improves various KGC methods. Specifically, the best variant of KGR3 achieves absolute Hits@1 improvements of 12.3% and 5.6% on the FB15k237 and WN18RR datasets.

We demonstrate numerical refocusing in coherent confocal laser scanning microscopy based on synthetic optical holography. This physics-based approach implements a computational propagation on the complex signal recovered in synthetic holography consistent with the wave physics and the parameters of the microscope. An experimental demonstration is shown to restore an in-focus image of a test object from data acquired at several focal plane off-sets. Numerical refocusing can provide focused views on samples with large height variation with a potential application in confocal optical surface profiling., Comment: 5 pages, 4 figures. \c{opyright} 2023 Optica Publishing Group. One print or electronic copy may be made for personal use only. Systematic reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modifications of the content of this paper are prohibited

As urbanization and climate change progress, urban heat becomes a priority for climate adaptation efforts. High temperatures concentrated in urban heat can drive increased risk of heat-related death and illness as well as increased energy demand for cooling. However, estimating the effects of urban heat is an ongoing field of research typically burdened by an imprecise description of the built environment, significant computational cost, and a lack of high-resolution estimates of the impacts of climate change. Here, we present open-source, computationally efficient machine learning methods that can improve the accuracy of urban temperature estimates when compared to historical reanalysis data. These models are applied to residential buildings in Los Angeles, and we compare the energy benefits of heat mitigation strategies to the impacts of climate change. We find that cooling demand is likely to increase substantially through midcentury, but engineered high-albedo surfaces could lessen this increase by more than 50%. The corresponding increase in heating demand complicates this narrative, but total annual energy use from combined heating and cooling with electric heat pumps in the Los Angeles urban climate is shown to benefit from the engineered cooling strategies under both current and future climates.

We present a sensor-agnostic spectral transformer as the basis for spectral foundation models. To that end, we introduce a Universal Spectral Representation (USR) that leverages sensor meta-data, such as sensing kernel specifications and sensing wavelengths, to encode spectra obtained from any spectral instrument into a common representation, such that a single model can ingest data from any sensor. Furthermore, we develop a methodology for pre-training such models in a self-supervised manner using a novel random sensor-augmentation and reconstruction pipeline to learn spectral features independent of the sensing paradigm. We demonstrate that our architecture can learn sensor independent spectral features that generalize effectively to sensors not seen during training. This work sets the stage for training foundation models that can both leverage and be effective for the growing diversity of spectral data.

Rare-earth ions are characterised by transitions with very narrow linewidths even in solid state crystals. Exceedingly long coherence times have been shown on both spin and optical transitions of rare-earth-ion doped crystals. A key factor, and generally the limitation, for such coherence times, is the effects of electronic and nuclear spins in the host crystal. Despite the attractive prospect, a low-strain, spin-free host crystal for rare-earth-ion dopants has not yet been demonstrated. The dopants experience the lowest strain when they substitute for another rare earth (including yttrium). However every stable isotope of the trivalent rare earth ions has either an electron spin, an nuclear spin, or both. The long optical coherence times reported here with erbium dopants in antiferromagnetically ordered gadolinium vanandate suggest an alternative method to achieve the quiet magnetic environment needed for long coherence times: use a magnetic host fully concentrated in electron spins and operate at temperatures low enough for these spins to be ordered. We also observe avoided crossings in the optical spectra, caused by strong coupling between the erbium ions and gadolinium magnons in the host crystal. This suggests the exciting prospect of microwave to optical quantum transduction using the rare-earth ions in these materials mediated by magnons of the host spins.

We discuss modular domain walls and gravitational waves in a class of supersymmetric models where quark and lepton flavour symmetry emerges from modular symmetry. In such models a single modulus field $\tau$ is often assumed to be stabilised at or near certain fixed point values such as $\tau = {\rm i}$ and $\tau = \omega$ (the cube root of unity), in its fundamental domain. We show that, in the global supersymmetry limit of certain classes of potentials, the vacua at these fixed points may be degenerate, leading to the formation of modular domain walls in the early Universe. Taking supergravity effects into account, in the background of a fixed dilaton field $S$, the degeneracy may be lifted, leading to a bias term in the potential allowing the domain walls to collapse. We study the resulting gravitational wave spectra arising from the dynamics of such modular domain walls, and assess their observability by current and future experiments, as a window into modular flavour symmetry., Comment: 23 pages, 6 figures

Run-time steering strategies like Medprompt are valuable for guiding large language models (LLMs) to top performance on challenging tasks. Medprompt demonstrates that a general LLM can be focused to deliver state-of-the-art performance on specialized domains like medicine by using a prompt to elicit a run-time strategy involving chain of thought reasoning and ensembling. OpenAI's o1-preview model represents a new paradigm, where a model is designed to do run-time reasoning before generating final responses. We seek to understand the behavior of o1-preview on a diverse set of medical challenge problem benchmarks. Following on the Medprompt study with GPT-4, we systematically evaluate the o1-preview model across various medical benchmarks. Notably, even without prompting techniques, o1-preview largely outperforms the GPT-4 series with Medprompt. We further systematically study the efficacy of classic prompt engineering strategies, as represented by Medprompt, within the new paradigm of reasoning models. We found that few-shot prompting hinders o1's performance, suggesting that in-context learning may no longer be an effective steering approach for reasoning-native models. While ensembling remains viable, it is resource-intensive and requires careful cost-performance optimization. Our cost and accuracy analysis across run-time strategies reveals a Pareto frontier, with GPT-4o representing a more affordable option and o1-preview achieving state-of-the-art performance at higher cost. Although o1-preview offers top performance, GPT-4o with steering strategies like Medprompt retains value in specific contexts. Moreover, we note that the o1-preview model has reached near-saturation on many existing medical benchmarks, underscoring the need for new, challenging benchmarks. We close with reflections on general directions for inference-time computation with LLMs., Comment: 25 pages

Topological data analysis (TDA) aims to extract noise-robust features from a data set by examining the number and persistence of holes in its topology. We show that a computational problem closely related to a core task in TDA -- determining whether a given hole persists across different length scales -- is $\mathsf{BQP}_1$-hard and contained in $\mathsf{BQP}$. This result implies an exponential quantum speedup for this problem under standard complexity-theoretic assumptions. Our approach relies on encoding the persistence of a hole in a variant of the guided sparse Hamiltonian problem, where the guiding state is constructed from a harmonic representative of the hole., Comment: 21 pages

Microkinetic models are key for evaluating industrial processes' efficiency and chemicals' environmental impact. Manual construction of these models is difficult and time-consuming, prompting a shift to automated methods. This study introduces SiMBA (Simplest Mechanism Builder Algorithm), a novel approach for generating microkinetic models from kinetic data. SiMBA operates through four phases: mechanism generation, mechanism translation, parameter estimation, and model comparison. Our approach systematically proposes reaction mechanisms, using matrix representations and a parallelized backtracking algorithm to manage complexity. These mechanisms are then translated into microkinetic models represented by ordinary differential equations, and optimized to fit available data. Models are compared using information criteria to balance accuracy and complexity, iterating until convergence to an optimal model is reached. Case studies on an aldol condensation reaction, and the dehydration of fructose demonstrate SiMBA's effectiveness in distilling complex kinetic behaviors into simple yet accurate models. While SiMBA predicts intermediates correctly for all case studies, it does not chemically identify intermediates, requiring expert input for complex systems. Despite this, SiMBA significantly enhances mechanistic exploration, offering a robust initial mechanism that accelerates the development and modeling of chemical processes.

Cardiovascular diseases (CVDs) are currently the leading cause of death worldwide, highlighting the critical need for early diagnosis and treatment. Machine learning (ML) methods can help diagnose CVDs early, but their performance relies on access to substantial data with high quality. However, the sensitive nature of healthcare data often restricts individual clinical institutions from sharing data to train sufficiently generalized and unbiased ML models. Federated Learning (FL) is an emerging approach, which offers a promising solution by enabling collaborative model training across multiple participants without compromising the privacy of the individual data owners. However, to the best of our knowledge, there has been limited prior research applying FL to the cardiovascular disease domain. Moreover, existing FL benchmarks and datasets are typically simulated and may fall short of replicating the complexity of natural heterogeneity found in realistic datasets that challenges current FL algorithms. To address these gaps, this paper presents the first real-world FL benchmark for cardiovascular disease detection, named FedCVD. This benchmark comprises two major tasks: electrocardiogram (ECG) classification and echocardiogram (ECHO) segmentation, based on naturally scattered datasets constructed from the CVD data of seven institutions. Our extensive experiments on these datasets reveal that FL faces new challenges with real-world non-IID and long-tail data. The code and datasets of FedCVD are available https://github.com/SMILELab-FL/FedCVD., Comment: 10 pages, 4 figures

Discrete representations of speech, obtained from Self-Supervised Learning (SSL) foundation models, are widely used, especially where there are limited data for the downstream task, such as for a low-resource language. Typically, discretization of speech into a sequence of symbols is achieved by unsupervised clustering of the latents from an SSL model. Our study evaluates whether discrete symbols - found using k-means - adequately capture tone in two example languages, Mandarin and Yoruba. We compare latent vectors with discrete symbols, obtained from HuBERT base, MandarinHuBERT, or XLS-R, for vowel and tone classification. We find that using discrete symbols leads to a substantial loss of tone information, even for language-specialised SSL models. We suggest that discretization needs to be task-aware, particularly for tone-dependent downstream tasks., Comment: Submitted to ICASSP 2025

Fine-tuning Large Language Models (LLMs) has become a crucial technique for adapting pre-trained models to downstream tasks. However, the enormous size of LLMs poses significant challenges in terms of computational complexity and resource requirements. Low-Rank Adaptation (LoRA) has emerged as a promising solution. However, there exists a gap between the practical performance of low-rank adaptations and its theoretical optimum. In this work, we propose eXtreme Gradient Boosting LoRA (XGBLoRA), a novel framework that bridges this gap by leveraging the power of ensemble learning. Inspired by gradient boosting, XGBLoRA iteratively learns and merges a sequence of LoRA adaptations to refine model predictions. It achieves better performance than the standard LoRA, while enjoying the computational efficiency of rank-1 adaptations. We provide theoretical analysis to show the convergence and optimality of our approach, and conduct extensive experiments on a range of natural language processing tasks. The results demonstrate that XGBLoRA consistently outperforms standard LoRA and achieves performance comparable to full fine-tuning with significantly fewer trainable parameters. This work advances parameter-efficient fine-tuning for LLMs, and offers a promising solution for adapting LLMs to downstream tasks while optimizing performance and efficiency., Comment: 19 pages

Inductive knowledge graph completion (KGC) aims to predict missing triples with unseen entities. Recent works focus on modeling reasoning paths between the head and tail entity as direct supporting evidence. However, these methods depend heavily on the existence and quality of reasoning paths, which limits their general applicability in different scenarios. In addition, we observe that latent type constraints and neighboring facts inherent in KGs are also vital in inferring missing triples. To effectively utilize all useful information in KGs, we introduce CATS, a novel context-aware inductive KGC solution. With sufficient guidance from proper prompts and supervised fine-tuning, CATS activates the strong semantic understanding and reasoning capabilities of large language models to assess the existence of query triples, which consist of two modules. First, the type-aware reasoning module evaluates whether the candidate entity matches the latent entity type as required by the query relation. Then, the subgraph reasoning module selects relevant reasoning paths and neighboring facts, and evaluates their correlation to the query triple. Experiment results on three widely used datasets demonstrate that CATS significantly outperforms state-of-the-art methods in 16 out of 18 transductive, inductive, and few-shot settings with an average absolute MRR improvement of 7.2%.

A recent paper (King, 2024) suggested that emission from the central supermassive black holes in high-redshift galaxies must be tightly collimated by the effects of partly expelling a super-Eddington mass supply. I show here that this idea predicts that these galaxies should produce very little detectable rest-frame X-ray emission, appear Compton thick, and show no easily detectable sign of outflows. All of these properties agree with current observations. To produce these effects, the mass supply to the black holes should exceed the Eddington rate by factors 50 - 100, which appears in line with conditions during the early growth of the holes. I note that theoretical derivations of the ratio of black hole mass to host galaxy stellar mass already predict that this should increase significantly at high redshift, in line with recent observations., Comment: 3 pages, accepted for MNRAS

This paper introduces Moonshine, a family of speech recognition models optimized for live transcription and voice command processing. Moonshine is based on an encoder-decoder transformer architecture and employs Rotary Position Embedding (RoPE) instead of traditional absolute position embeddings. The model is trained on speech segments of various lengths, but without using zero-padding, leading to greater efficiency for the encoder during inference time. When benchmarked against OpenAI's Whisper tiny-en, Moonshine Tiny demonstrates a 5x reduction in compute requirements for transcribing a 10-second speech segment while incurring no increase in word error rates across standard evaluation datasets. These results highlight Moonshine's potential for real-time and resource-constrained applications., Comment: 7 pages, 6 figures, 3 tables

Monitoring of Diffuse Intrinsic Pontine Glioma (DIPG) and Diffuse Midline Glioma (DMG) brain tumors in pediatric patients is key for assessment of treatment response. Response Assessment in Pediatric Neuro-Oncology (RAPNO) guidelines recommend the volumetric measurement of these tumors using MRI. Segmentation challenges, such as the Brain Tumor Segmentation (BraTS) Challenge, promote development of automated approaches which are replicable, generalizable and accurate, to aid in these tasks. The current study presents a novel adaptation of existing nnU-Net approaches for pediatric brain tumor segmentation, submitted to the BraTS-PEDs 2024 challenge. We apply an adapted nnU-Net with hierarchical cascades to the segmentation task of the BraTS-PEDs 2024 challenge. The residual encoder variant of nnU-Net, used as our baseline model, already provides high quality segmentations. We incorporate multiple changes to the implementation of nnU-Net and devise a novel two-stage cascaded nnU-Net to segment the substructures of brain tumors from coarse to fine. Using outputs from the nnU-Net Residual Encoder (trained to segment CC, ED, ET and NET tumor labels from T1w, T1w-CE, T2w and T2-FLAIR MRI), these are passed to two additional models one classifying ET versus NET and a second classifying CC vs ED using cascade learning. We use radiological guidelines to steer which multi parametric MRI (mpMRI) to use in these cascading models. Compared to a default nnU-Net and an ensembled nnU-net as baseline approaches, our novel method provides robust segmentations for the BraTS-PEDs 2024 challenge, achieving mean Dice scores of 0.657, 0.904, 0.703, and 0.967, and HD95 of 76.2, 10.1, 111.0, and 12.3 for the ET, NET, CC and ED, respectively.

We consider the spreading dynamics of the Fisher-KPP equation in a shifting environment, by analyzing the limit of the rate function of the solutions. For environments with a weak monotone condition, it was demonstrated in a previous paper that the rate function converges to the unique Ishii solution of the underlying Hamilton-Jacobi equations. In case the environment does not satisfy the weak monotone condition, we show that the rate function is then characterized by the Hamilton-Jacobi equation with a dynamic junction condition, which depends additionally on the generalized eigenvalue derived from the environmental function. Our results applies to the case when the environment has multiple shifting speeds, and clarify the connection with previous results on nonlocally pulled fronts and forced traveling waves.

To many Chomsky's debates with Quine and Skinner are an updated version of the Rationalist Empiricist debates of the 17th century. The consensus being that Chomsky's Rationalism was victorious. This dispute has reemerged with the advent of Large Language Models. With some arguing that LLMs vindicate rationalism because of the necessity of building in innate biases to make them work. The necessity of building in innate biases is taken to prove that empiricism hasn't got the conceptual resources to explain linguistic competence. Such claims depend on the nature of the empiricism one is endorsing. Externalized Empiricism has no difficulties with innate apparatus once they are determined empirically (Quine 1969). Thus, externalized empiricism is not refuted because of the need to build in innate biases in LLMs. Furthermore, the relevance of LLMs to the rationalist empiricist debate in relation to humans is dubious. For any claim about whether LLMs learn in an empiricist manner to be relevant to humans it needs to be shown that LLMs and humans learn in the same way. Two key features distinguish humans and LLMs. Humans learn despite a poverty of stimulus and LLMs learn because of an incredibly rich stimulus. Human linguistic outputs are grounded in sensory experience and LLMs are not. These differences in how the two learn indicates that they both use different underlying competencies to produce their output. Therefore, any claims about whether LLMs learn in an empiricist manner are not relevant to whether humans learn in an empiricist manner.

We carried out a comprehensive crystal structure characterization of Ti-doped lithium ruthenate (Li$_2$Ti$_x$Ru$_{1-x}$O$_3$), to investigate the effect of Ti-doping on the structural phase transition. Experimental tools sensitive to the average structure (X-ray diffraction), as well as those sensitive to local structure (Extended X-ray Absorption Fine Structure, EXAFS; pair distribution function, PDF) are used. We observed non-monotonic dependence of the structural transition temperature on the Ti-doping level. At low doping, the transition temperature slightly increases with doping, while at high doping, the temperature decreases significantly with doping. We note two important observations from our studies. First, Ti K-edge EXAFS data shows persistent Ti-Ru dimerization even with substantial Ti doping. Second, we were able to use the PDF data to estimate the dimer correlation length above the transition temperature, which would correspond to the size of the proposed local `dimer clusters' formed by Ru-Ru and Ti-Ru neighbours. The dimer correlation length is found to be around 10~\AA, which remains robust regardless of doping. Our study therefore suggests that Ti$^{4+}$ with its $d^0$ electronic configuration is a special type of dopant when replacing Ru., Comment: 12 pages, 11 figures

Efficiently retrieving and synthesizing information from large-scale multimodal collections has become a critical challenge. However, existing video retrieval datasets suffer from scope limitations, primarily focusing on matching descriptive but vague queries with small collections of professionally edited, English-centric videos. To address this gap, we introduce $\textbf{MultiVENT 2.0}$, a large-scale, multilingual event-centric video retrieval benchmark featuring a collection of more than 218,000 news videos and 3,906 queries targeting specific world events. These queries specifically target information found in the visual content, audio, embedded text, and text metadata of the videos, requiring systems leverage all these sources to succeed at the task. Preliminary results show that state-of-the-art vision-language models struggle significantly with this task, and while alternative approaches show promise, they are still insufficient to adequately address this problem. These findings underscore the need for more robust multimodal retrieval systems, as effective video retrieval is a crucial step towards multimodal content understanding and generation tasks.

Machine learning has revolutionized the modeling of clinical timeseries data. Using machine learning, a Deep Neural Network (DNN) can be automatically trained to learn a complex mapping of its input features for a desired task. This is particularly valuable in Electronic Health Record (EHR) databases, where patients often spend extended periods in intensive care units (ICUs). Machine learning serves as an efficient method for extract meaningful information. However, many state-of-the-art (SOTA) methods for training DNNs demand substantial volumes of labeled data, posing significant challenges for clinics in terms of cost and time. Self-supervised learning offers an alternative by allowing practitioners to extract valuable insights from data without the need for costly labels. Yet, current SOTA methods often necessitate large data batches to achieve optimal performance, increasing computational demands. This presents a challenge when working with long clinical timeseries data. To address this, we propose an efficient method of contrastive pretraining tailored for long clinical timeseries data. Our approach utilizes an estimator for negative pair comparison, enabling effective feature extraction. We assess the efficacy of our pretraining using standard self-supervised tasks such as linear evaluation and semi-supervised learning. Additionally, our model demonstrates the ability to impute missing measurements, providing clinicians with deeper insights into patient conditions. We demonstrate that our pretraining is capable of achieving better performance as both the size of the model and the size of the measurement vocabulary scale. Finally, we externally validate our model, trained on the MIMIC-III dataset, using the eICU dataset. We demonstrate that our model is capable of learning robust clinical information that is transferable to other clinics.

Panel data arise when time series measurements are collected from multiple, dynamically independent but structurally related systems. In such cases, each system's time series can be modeled as a partially observed Markov process (POMP), and the ensemble of these models is called a PanelPOMP. If the time series are relatively short, statistical inference for each time series must draw information from across the entire panel. Every time series has a name, called its unit label, which may correspond to an object on which that time series was collected. Differences between units may be of direct inferential interest or may be a nuisance for studying the commonalities. The R package panelPomp supports analysis of panel data via a general class of PanelPOMP models. This includes a suite of tools for manipulation of models and data that take advantage of the panel structure. The panelPomp package currently emphasizes recent advances enabling likelihood-based inference via simulation-based algorithms. However, the general framework provided by panelPomp supports development of additional, new inference methodology for panel data.

While traditional earphones primarily offer private audio spaces, so-called "earables" emerged to offer a variety of sensing capabilities. Pioneering platforms like OpenEarable have introduced novel sensing platforms targeted at the ears, incorporating various sensors. The proximity of the ears to the eyes, brain, and facial muscles has also sparked investigation into sensing biopotentials. However, currently there is no platform available that is targeted at the ears to sense biopotentials. To address this gap, we introduce OpenEarable ExG - an open-source hardware platform designed to measure biopotentials in and around the ears. OpenEarable ExG can be freely configured and has up to 7 sensing channels. We initially validate OpenEarable ExG in a study with a left-right in-ear dual-electrode montage setup with 3 participants. Our results demonstrate the successful detection of smooth pursuit eye movements via Electrooculography (EOG), alpha brain activity via Electroencephalography (EEG), and jaw clenching via Electromyography (EMG). OpenEarable ExG is part of the OpenEarable initiative and is fully open-source under MIT license., Comment: 5 pages, 3 figures

Nano-patterned magnetic materials have opened new venues on the investigation of strongly correlated phenomena including artificial spin-ice systems, geometric frustration, magnetic monopoles, for technologically important applications such as reconfigurable ferromagnetism. With the advent of atomically thin two-dimensional (2D) van der Waals (vdW) magnets a pertinent question is whether such compounds could make their way into this realm where interactions can be tailored so that unconventional states of matter could be assessed. Here we show that square islands of CrGeTe3 vdW ferromagnets distributed in a grid manifest antiferromagnetic correlations, essential to enable frustration resulting in an artificial spin-ice. By using a combination of SQUID-on-tip microscopy, focused ion beam lithography, and atomistic spin dynamic simulations, we show that pristine, isolated CGT flakes as small as 150*150*60 nm3 have tunable dipole-dipole interactions, which can be precisely controlled by their lateral spacing. There is a crossover between non-interacting islands and significant inter-island anticorrelation depending how they are spatially distributed allowing the creation of complex magnetic patterns not observable at the isolated flakes. Our findings suggest that the cross-talk between the nano-patterned magnets can be explored in the generation of even more complex spin configurations where exotic interactions may be manipulated in an unprecedent way., Comment: Main text: 12 pages, 5 figures. Supplementary information: 11 pages, 5 figures

FAMU is an INFN-led muonic atom physics experiment based at the RIKEN-RAL muon facility at the ISIS Neutron and Muon Source (United Kingdom). The aim of FAMU is to measure the hyperfine splitting in muonic hydrogen to determine the value of the proton Zemach radius with accuracy better than 1%.The experiment has a scintillating-fibre hodoscope for beam monitoring and data normalisation. In order to carry out muon flux estimation, low-rate measurements were performed to extract the single-muon average deposited charge. Then, detector simulation in Geant4 and FLUKA allowed a thorough understanding of the single-muon response function, crucial for determining the muon flux. This work presents the design features of the FAMU beam monitor, along with the simulation and absolute calibration measurements in order to enable flux determination and enable data normalisation.

Continual learning (CL) aims to empower machine learning models to learn continually from new data, while building upon previously acquired knowledge without forgetting. As machine learning models have evolved from small to large pre-trained architectures, and from supporting unimodal to multimodal data, multimodal continual learning (MMCL) methods have recently emerged. The primary challenge of MMCL is that it goes beyond a simple stacking of unimodal CL methods, as such straightforward approaches often yield unsatisfactory performance. In this work, we present the first comprehensive survey on MMCL. We provide essential background knowledge and MMCL settings, as well as a structured taxonomy of MMCL methods. We categorize existing MMCL methods into four categories, i.e., regularization-based, architecture-based, replay-based, and prompt-based methods, explaining their methodologies and highlighting their key innovations. Additionally, to prompt further research in this field, we summarize open MMCL datasets and benchmarks, and discuss several promising future directions for investigation and development. We have also created a GitHub repository for indexing relevant MMCL papers and open resources available at https://github.com/LucyDYu/Awesome-Multimodal-Continual-Learning.

For nearly a century and a half the Stokes phenomenon had been perceived as a discontinuous change in the asymptotic representation of a function. In 1989 Berry demonstrated how it is possible to smooth out this discontinuity in broad classes of problems with the prefactor for the exponentially small contribution that is being switched on/off taking the universal form of an error function. Following pioneering work of Berk {\it et al.} \cite{BNR82} and the Japanese school of formally exact asymptotics \cite{Aokietal1994,AKT01}, the concept of the higher-order Stokes phenomenon was introduced in \cite{HLO04} and \cite{CM05}, whereby the ability for the exponentially small terms to cause a Stokes phenomenon may change, depending on the values of parameters in the problem, corresponding to the associated singularities in the Borel plane transitioning between different Riemann sheets. Until now, the higher-order Stokes phenomenon has also been treated as a discontinuous event. In this paper we show how the higher-order Stokes phenomenon is, in fact, also smooth and occurs universally with a prefactor that takes the form of a new special function, based on a Gaussian convolution of an error function that gives rise to a rich structure. We provide a rigorous derivation of the result, with examples spanning the gamma function, a second-order nonlinear ODE and the telegraph equation, giving rise to a ghost-like smooth contribution that is present in the vicinity of a Stokes line, but which rapidly tends to zero on either side. We also include a rigorous derivation and example of the effect of the smoothed higher-order Stokes phenomenon on the individual terms in the asymptotic series, where the additional contributions appear prefactored by an error function., Comment: 49 pages, 15 figures

We present projected walk on spheres (PWoS), a novel pointwise and discretization-free Monte Carlo solver for surface PDEs with Dirichlet boundaries, as a generalization of the walk on spheres method (WoS) [Muller 1956; Sawhney and Crane 2020]. We adapt the recursive relationship of WoS designed for PDEs in volumetric domains to a volumetric neighborhood around the surface, and at the end of each recursion step, we project the sample point on the sphere back to the surface. We motivate this simple modification to WoS with the theory of the closest point extension used in the closest point method. To define the valid volumetric neighborhood domain for PWoS, we develop strategies to estimate the local feature size of the surface and to compute the distance to the Dirichlet boundaries on the surface extended in their normal directions. We also design a mean value filtering method for PWoS to improve the method's efficiency when the surface is represented as a polygonal mesh or a point cloud. Finally, we study the convergence of PWoS and demonstrate its application to graphics tasks, including diffusion curves, geodesic distance computation, and wave propagation animation. We show that our method works with various types of surfaces, including a surface of mixed codimension., Comment: Accepted to SIGGRAPH Asia 2024 (Conference Papers). See https://rsugimoto.net/ProjectedWalkOnSpheres/ for updates

Large language models (LLMs) have demonstrated remarkable performance on various tasks. However, it remains an open question whether the default Euclidean space is the most suitable choice for embedding tokens in LLMs. In this study, we first investigate the non-Euclidean characteristics of LLMs. Our findings reveal that token frequency follows a power-law distribution, with high-frequency tokens clustering near the origin and low-frequency tokens positioned farther away. Additionally, token embeddings exhibit a high degree of hyperbolicity, indicating a latent tree-like structure in the embedding space. Building on the observation, we propose to efficiently fine-tune LLMs in hyperbolic space to better exploit the underlying complex structures. However, we found that this fine-tuning in hyperbolic space cannot be achieved with naive application of exponential and logarithmic maps, when the embedding and weight matrices both reside in Euclidean space. To address this technique issue, we introduce a new method called hyperbolic low-rank efficient fine-tuning, HypLoRA, that performs low-rank adaptation directly on the hyperbolic manifold, avoiding the cancellation effect caused by the exponential and logarithmic maps, thus preserving the hyperbolic modeling capabilities. Through extensive experiments, we demonstrate that HypLoRA significantly enhances the performance of LLMs on reasoning tasks, particularly for complex reasoning problems. In particular, HypLoRA improves the performance in the complex AQuA dataset by up to 13.0%, showcasing its effectiveness in handling complex reasoning challenges, Comment: The preliminary work was accepted for the ICML 2024 LLM Cognition Workshop, and this version includes new investigations, analyses, experiments, and results

A high-power laser pulse at normal incidence onto a plane solid target will generate odd harmonics of its frequency. However, the spacing of the harmonic lines in this configuration is fixed. Here, we study harmonic generation using two laser beams incident on a plane target at small, opposite angles to the target normal, via particle-in-cell simulations. When looking at the harmonic radiation in a specific direction via a narrow slit or pinhole, we select an angle-dependent subset of the harmonic spectrum. This way, we obtain a harmonic frequency comb that we control via the observation angle and the input laser frequency. The divergence of the harmonic radiation will be reduced by using wider laser spots, thus increasing the efficacy of the scheme. We will discuss extensions to this scheme, such as using beams with unequal frequencies, a slight tilt of the target, or employing more than two beams., Comment: 10 pages, 5 figures. This version provides a more detailed description of the laser pulses used in the simulations

Jupiter's Great Red Spot (GRS) was mapped by the James Webb Space Telescope (JWST)/Mid-Infrared Instrument (4.9-27.9 micron) in July and August 2022. These observations took place alongside a suite of visual and infrared observations from; Hubble, JWST/NIRCam, Very Large Telescope/VISIR and amateur observers which provided both spatial and temporal context across the jovian disc. The stratospheric temperature structure retrieved using the NEMESIS software revealed a series of hot-spots above the GRS. These could be the consequence of GRS-induced wave activity. In the troposphere, the temperature structure was used to derive the thermal wind structure of the GRS vortex. These winds were only consistent with the independently determined wind field by JWST/NIRCam at 240 mbar if the altitude of the Hubble-derived winds were located around 1,200 mbar, considerably deeper than previously assumed. No enhancement in ammonia was found within the GRS but a link between elevated aerosol and phosphine abundances was observed within this region. North-south asymmetries were observed in the retrieved temperature, ammonia, phosphine and aerosol structure, consistent with the GRS tilting in the north-south direction. Finally, a small storm was captured north-west of the GRS that displayed a considerable excess in retrieved phosphine abundance, suggestive of vigorous convection. Despite this, no ammonia ice was detected in this region. The novelty of JWST required us to develop custom-made software to resolve challenges in calibration of the data. This involved the derivation of the "FLT-5" wavelength calibration solution that has subsequently been integrated into the standard calibration pipeline., Comment: 53 pages, 19 figures, 4 tables

We present an adaptation of the exoplanet transit model code batman, in order to permit the generation of X-ray transits. Our underlying extended coronal model assumes an isothermal plasma that is radially symmetric. While this ignores the effect of bright, active regions, observations of transits in X-rays will require averaging across multiple epochs of data for the foreseeable future, significantly reducing the importance of more complex modelling. Our publicly available code successfully generates the predicted W-shaped transit profile in X-rays due to the optically thin nature of the emission, which concentrates the expected observational emission around the limb of the photospheric stellar disc. We provide some examples based on the best known X-ray transit target, HD 189733b, and examine the effect of varying the planet size, coronal temperature, and impact parameter on the resulting transit profile. We also derived scaling relationships for how the overall transit detectability is affected by changing these parameters. Over most of the parameter space, we find that the detectability scales linearly with the cross-sectional area of the planet in X-rays. The relationship with increasing coronal temperature is less fixed, but averages out to a power law with slope $-1/4$ except when the impact parameter is high. Indeed, varying impact parameter has little effect on detectability at all until it approaches unity., Comment: 16 pages, 8 figures, 2 tables. Accepted for publication in the Astronomical Journal. Code available at https://github.com/georgewking/batmanX-rays

Large Language Models (LLMs) have recently garnered significant attention, primarily for their capabilities in text-based interactions. However, natural human interaction often relies on speech, necessitating a shift towards voice-based models. A straightforward approach to achieve this involves a pipeline of ``Automatic Speech Recognition (ASR) + LLM + Text-to-Speech (TTS)", where input speech is transcribed to text, processed by an LLM, and then converted back to speech. Despite being straightforward, this method suffers from inherent limitations, such as information loss during modality conversion and error accumulation across the three stages. To address these issues, Speech Language Models (SpeechLMs) -- end-to-end models that generate speech without converting from text -- have emerged as a promising alternative. This survey paper provides the first comprehensive overview of recent methodologies for constructing SpeechLMs, detailing the key components of their architecture and the various training recipes integral to their development. Additionally, we systematically survey the various capabilities of SpeechLMs, categorize the evaluation metrics for SpeechLMs, and discuss the challenges and future research directions in this rapidly evolving field., Comment: Work in progress

Tin oxide aerogels were synthesized using an epoxide-assisted technique and characterized with Fourier transform infrared, X-ray diffraction, and UV-Vis to study the effects of post-synthesis annealing and peroxide treatment. While bulk tin oxide exhibits an optical bandgap of $3.6$ eV, its aerogel form often displays a larger apparent bandgap around $4.6$ eV due to defects. Our study reveals that annealing induces a partial phase change from SnO$_2$ to SnO, but is ineffective in removing defects. Conversely, peroxide passivation effectively lowers the bandgap and disorder levels, suggesting that dangling bonds are the primary cause of the increased bandgap in tin oxide aerogels. These findings offer insights for optimizing the optical properties of tin oxide aerogels for applications like solar cells.

Conversational search offers an easier and faster alternative to conventional web search, while having downsides like lack of source verification. Research has examined performance disparities between these two systems in different settings. However, little work has considered the effects of variations within a given search task. We hypothesize that psychological distance - one''s perceived closeness to a target event - affects information needs in search tasks, and investigate the corresponding effects on user preferences between web and conversational search systems. We find that with greater psychological distances, users perceive conversational search as more credible, useful, enjoyable, and easy to use, and demonstrate increased preference for this system. We reveal qualitative reasons for these differences and provide design implications for search system designers., Comment: 28 pages

Non-maximum suppression (NMS) is an indispensable post-processing step in object detection. With the continuous optimization of network models, NMS has become the ``last mile'' to enhance the efficiency of object detection. This paper systematically analyzes NMS from a graph theory perspective for the first time, revealing its intrinsic structure. Consequently, we propose two optimization methods, namely QSI-NMS and BOE-NMS. The former is a fast recursive divide-and-conquer algorithm with negligible mAP loss, and its extended version (eQSI-NMS) achieves optimal complexity of $\mathcal{O}(n\log n)$. The latter, concentrating on the locality of NMS, achieves an optimization at a constant level without an mAP loss penalty. Moreover, to facilitate rapid evaluation of NMS methods for researchers, we introduce NMS-Bench, the first benchmark designed to comprehensively assess various NMS methods. Taking the YOLOv8-N model on MS COCO 2017 as the benchmark setup, our method QSI-NMS provides $6.2\times$ speed of original NMS on the benchmark, with a $0.1\%$ decrease in mAP. The optimal eQSI-NMS, with only a $0.3\%$ mAP decrease, achieves $10.7\times$ speed. Meanwhile, BOE-NMS exhibits $5.1\times$ speed with no compromise in mAP.

This paper introduces the hypothetical track-length fitting algorithm, a novel method for measuring the kinetic energies of ionizing particles in liquid argon time projection chambers (LArTPCs). The algorithm finds the most probable offset in track length for a track-like object by comparing the measured ionization density as a function of position with a theoretical prediction of the energy loss as a function of the energy, including models of electron recombination and detector response. The algorithm can be used to measure the energies of particles that interact before they stop, such as charged pions that are absorbed by argon nuclei. The algorithm's energy measurement resolutions and fractional biases are presented as functions of particle kinetic energy and number of track hits using samples of stopping secondary charged pions in data collected by the ProtoDUNE-SP detector, and also in a detailed simulation. Additional studies describe impact of the dE/dx model on energy measurement performance. The method described in this paper to characterize the energy measurement performance can be repeated in any LArTPC experiment using stopping secondary charged pions.

This study aims to investigate whether GPT-4 can effectively grade assignments for design university students and provide useful feedback. In design education, assignments do not have a single correct answer and often involve solving an open-ended design problem. This subjective nature of design projects often leads to grading problems,as grades can vary between different raters,for instance instructor from engineering background or architecture background. This study employs an iterative research approach in developing a Custom GPT with the aim of achieving more reliable results and testing whether it can provide design students with constructive feedback. The findings include: First,through several rounds of iterations the inter-reliability between GPT and human raters reached a level that is generally accepted by educators. This indicates that by providing accurate prompts to GPT,and continuously iterating to build a Custom GPT, it can be used to effectively grade students' design assignments, serving as a reliable complement to human raters. Second, the intra-reliability of GPT's scoring at different times is between 0.65 and 0.78. This indicates that, with adequate instructions, a Custom GPT gives consistent results which is a precondition for grading students. As consistency and comparability are the two main rules to ensure the reliability of educational assessment, this study has looked at whether a Custom GPT can be developed that adheres to these two rules. We finish the paper by testing whether Custom GPT can provide students with useful feedback and reflecting on how educators can develop and iterate a Custom GPT to serve as a complementary rater., Comment: 25 pages, 5 figures

In this paper, we introduce MIO, a novel foundation model built on multimodal tokens, capable of understanding and generating speech, text, images, and videos in an end-to-end, autoregressive manner. While the emergence of large language models (LLMs) and multimodal large language models (MM-LLMs) propels advancements in artificial general intelligence through their versatile capabilities, they still lack true any-to-any understanding and generation. Recently, the release of GPT-4o has showcased the remarkable potential of any-to-any LLMs for complex real-world tasks, enabling omnidirectional input and output across images, speech, and text. However, it is closed-source and does not support the generation of multimodal interleaved sequences. To address this gap, we present MIO, which is trained on a mixture of discrete tokens across four modalities using causal multimodal modeling. MIO undergoes a four-stage training process: (1) alignment pre-training, (2) interleaved pre-training, (3) speech-enhanced pre-training, and (4) comprehensive supervised fine-tuning on diverse textual, visual, and speech tasks. Our experimental results indicate that MIO exhibits competitive, and in some cases superior, performance compared to previous dual-modal baselines, any-to-any model baselines, and even modality-specific baselines. Moreover, MIO demonstrates advanced capabilities inherent to its any-to-any feature, such as interleaved video-text generation, chain-of-visual-thought reasoning, visual guideline generation, instructional image editing, etc., Comment: Technical Report. Codes and models are available in https://github.com/MIO-Team/MIO