Search

Superconducting Digital (SCD) technology offers significant potential for enhancing the performance of next generation large scale compute workloads. By leveraging advanced lithography and a 300 mm platform, SCD devices can reduce energy consumption and boost computational power. This paper presents a cross-layer modeling approach to evaluate the system-level performance benefits of SCD architectures for Large Language Model (LLM) training and inference. Our findings, based on experimental data and Pulse Conserving Logic (PCL) design principles, demonstrate substantial performance gain in both training and inference. We are, thus, able to convincingly show that the SCD technology can address memory and interconnect limitations of present day solutions for next-generation compute systems., Comment: 8 figures

We define a Grothendieck ring of varieties for log schemes. It is generated by one additional class ``$P$'' over the usual Grothendieck ring. We show the na\"ive definition of log Hodge numbers does not make sense for all log schemes. We offer an alternative that does., Comment: 19 pages. Comments welcome

In this paper we study 2D Fourier expansions for a general class of planar measures $\mu$, generally singular, but assumed compactly supported in $\mathbb{R}^2$. We focus on the following question: When does $L^2(\mu)$ admit a 2D system of Fourier expansions? We offer concrete conditions allowing an affirmative answer to the question for a large class of Borel probability measures, and we present an explicit Fourier duality for these cases. Our 2D Fourier analysis relies on a detailed conditioning-analysis. For a given $\mu$, it is based on the corresponding systems of 1D measures consisting of a marginal measure and associated family of conditional measures computed from $\mu$ by the Rokhlin Disintegration Theorem. Our identified $L^2(\mu)$-Fourier expansions are special in two ways: For our measures $\mu$, the Fourier expansions are generally non-orthogonal, but nonetheless, they lend themselves to algorithmic computations. Second, we further stress that our class of 2D measures $\mu$ considered here go beyond what exists in the literature. In particular, our measures do not require affine iterated function system (IFS) properties, but we do study grid IFS measures in detail and provide some technical criteria guaranteeing their admission of Fourier expansions. Our analyses make use of estimates for the Hausdorff dimensions of the measure supports. An important class of examples addressed in this paper is fractal Bedford-McMullen carpets.

The rise of Large Language Models (LLMs) has revolutionized natural language processing across numerous languages and tasks. However, evaluating LLM performance in a consistent and meaningful way across multiple European languages remains challenging, especially due to the scarcity of language-parallel multilingual benchmarks. We introduce a multilingual evaluation approach tailored for European languages. We employ translated versions of five widely-used benchmarks to assess the capabilities of 40 LLMs across 21 European languages. Our contributions include examining the effectiveness of translated benchmarks, assessing the impact of different translation services, and offering a multilingual evaluation framework for LLMs that includes newly created datasets: EU20-MMLU, EU20-HellaSwag, EU20-ARC, EU20-TruthfulQA, and EU20-GSM8K. The benchmarks and results are made publicly available to encourage further research in multilingual LLM evaluation.

This paper presents a comprehensive overview of the data preparation pipeline developed for the OpenGPT-X project, a large-scale initiative aimed at creating open and high-performance multilingual large language models (LLMs). The project goal is to deliver models that cover all major European languages, with a particular focus on real-world applications within the European Union. We explain all data processing steps, starting with the data selection and requirement definition to the preparation of the final datasets for model training. We distinguish between curated data and web data, as each of these categories is handled by distinct pipelines, with curated data undergoing minimal filtering and web data requiring extensive filtering and deduplication. This distinction guided the development of specialized algorithmic solutions for both pipelines. In addition to describing the processing methodologies, we provide an in-depth analysis of the datasets, increasing transparency and alignment with European data regulations. Finally, we share key insights and challenges faced during the project, offering recommendations for future endeavors in large-scale multilingual data preparation for LLMs.

This work is devoted to the stochastic Zakharov system in dimension four, which is the energy-critical dimension. First, we prove local well-posedness in the energy space $H^1\times L^2$ up to the maximal existence time and a blow-up alternative. Second, we prove that for large data solutions exist globally as long as energy and wave mass are below the ground state threshold. Third, we prove a regularization by noise phenomenon: the probability of global existence and scattering goes to one if the strength of the (non-conservative) noise goes to infinity. The proof is based on the refined rescaling approach and a new functional framework, where both Fourier restriction and local smoothing norms are used as well as a (uniform) double endpoint Strichartz and local smoothing inequality for the Schr\"odinger equation with certain rough and time dependent lower order perturbations., Comment: 56 pages

We present two multilingual LLMs designed to embrace Europe's linguistic diversity by supporting all 24 official languages of the European Union. Trained on a dataset comprising around 60% non-English data and utilizing a custom multilingual tokenizer, our models address the limitations of existing LLMs that predominantly focus on English or a few high-resource languages. We detail the models' development principles, i.e., data composition, tokenizer optimization, and training methodologies. The models demonstrate competitive performance across multilingual benchmarks, as evidenced by their performance on European versions of ARC, HellaSwag, MMLU, and TruthfulQA.

The cell is arguably the most fundamental unit of life and is central to understanding biology. Accurate modeling of cells is important for this understanding as well as for determining the root causes of disease. Recent advances in artificial intelligence (AI), combined with the ability to generate large-scale experimental data, present novel opportunities to model cells. Here we propose a vision of leveraging advances in AI to construct virtual cells, high-fidelity simulations of cells and cellular systems under different conditions that are directly learned from biological data across measurements and scales. We discuss desired capabilities of such AI Virtual Cells, including generating universal representations of biological entities across scales, and facilitating interpretable in silico experiments to predict and understand their behavior using virtual instruments. We further address the challenges, opportunities and requirements to realize this vision including data needs, evaluation strategies, and community standards and engagement to ensure biological accuracy and broad utility. We envision a future where AI Virtual Cells help identify new drug targets, predict cellular responses to perturbations, as well as scale hypothesis exploration. With open science collaborations across the biomedical ecosystem that includes academia, philanthropy, and the biopharma and AI industries, a comprehensive predictive understanding of cell mechanisms and interactions has come into reach.

We construct a structure preserving non-conforming finite element approximation scheme for the bi-harmonic wave maps into spheres equation. It satisfies a discrete energy law and preserves the non-convex sphere constraint of the continuous problem. The discrete sphere constraint is enforced at the mesh-points via a discrete Lagrange multiplier. This approach restricts the spatial approximation to the (non-conforming) linear finite elements. We show that the numerical approximation converges to the weak solution of the continuous problem in spatial dimension $d=1$. The convergence analysis in dimensions $d>1$ is complicated by the lack of a discrete product rule as well as the low regularity of the numerical approximation in the non-conforming setting. Hence, we show convergence of the numerical approximation in higher-dimensions by introducing additional stabilization terms in the numerical approximation. We present numerical experiments to demonstrate the performance of the proposed numerical approximation and to illustrate the regularizing effect of the bi-Laplacian which prevents the formation of singularities., Comment: 18 pages, 8 figures

Self-injection locking (SIL) of laser diodes to microresonators is a powerful technique that enables compact narrow linewidth lasers. Here, we extend this technique to chip-integrated Fabry-Perot microresonators, which offer high-quality factors and large mode volumes in a compact footprint, reducing fundamental thermo-refractive noise. The resonators consist of a silicon nitride waveguide terminated by two photonic crystal reflectors fabricated via scalable ultraviolet lithography. Evanescent side-coupling allows precise tuning of the SIL feedback mechanism. We present a theoretical model and experimentally demonstrate SIL, resulting in a fundamental thermorefractive-noise-limited laser. We achieve excellent agreement between the experimental results and theoretical model. These results complement current SIL techniques and are relevant to chip-scale low-noise laser systems., Comment: 4 pages, 4 figures

Recent advancements in Large Language Models (LLMs) have showcased their proficiency in answering natural language queries. However, their effectiveness is hindered by limited domain-specific knowledge, raising concerns about the reliability of their responses. We introduce a hybrid system that augments LLMs with domain-specific knowledge graphs (KGs), thereby aiming to enhance factual correctness using a KG-based retrieval approach. We focus on a medical KG to demonstrate our methodology, which includes (1) pre-processing, (2) Cypher query generation, (3) Cypher query processing, (4) KG retrieval, and (5) LLM-enhanced response generation. We evaluate our system on a curated dataset of 69 samples, achieving a precision of 78\% in retrieving correct KG nodes. Our findings indicate that the hybrid system surpasses a standalone LLM in accuracy and completeness, as verified by an LLM-as-a-Judge evaluation method. This positions the system as a promising tool for applications that demand factual correctness and completeness, such as target identification -- a critical process in pinpointing biological entities for disease treatment or crop enhancement. Moreover, its intuitive search interface and ability to provide accurate responses within seconds make it well-suited for time-sensitive, precision-focused research contexts. We publish the source code together with the dataset and the prompt templates used., Comment: 10 pages, 7 figures

We extend Bourgain's $L^2$-wellposedness result for the KP-II equation on $\mathbb{T}^2$ to initial data with negative Sobolev regularity. The key ingredient is a new linear $L^4$-Strichartz estimate which is effective on frequency-dependent time scales. The $L^4$-Strichartz estimates follow from combining an $\ell^2$-decoupling inequality recently proved by Guth--Maldague--Oh with semiclassical Strichartz estimates. Moreover, we rely on a variant of Bourgain's bilinear Strichartz estimate on frequency-dependent times, which is proved via the C\'ordoba--Fefferman square function estimate., Comment: 60 pages

While preliminary findings indicate that multilingual LLMs exhibit reduced bias compared to monolingual ones, a comprehensive understanding of the effect of multilingual training on bias mitigation, is lacking. This study addresses this gap by systematically training six LLMs of identical size (2.6B parameters) and architecture: five monolingual models (English, German, French, Italian, and Spanish) and one multilingual model trained on an equal distribution of data across these languages, all using publicly available data. To ensure robust evaluation, standard bias benchmarks were automatically translated into the five target languages and verified for both translation quality and bias preservation by human annotators. Our results consistently demonstrate that multilingual training effectively mitigates bias. Moreover, we observe that multilingual models achieve not only lower bias but also superior prediction accuracy when compared to monolingual models with the same amount of training data, model architecture, and size., Comment: 19 pages, 8 figures, C3NLP 2024

Large Language Models (LLMs) have been increasingly used in real-world settings, yet their strategic decision-making abilities remain largely unexplored. To fully benefit from the potential of LLMs, it's essential to understand their ability to function in complex social scenarios. Game theory, which is already used to understand real-world interactions, provides a good framework for assessing these abilities. This work investigates the performance and merits of LLMs in canonical game-theoretic two-player non-zero-sum games, Stag Hunt and Prisoner Dilemma. Our structured evaluation of GPT-3.5, GPT-4-Turbo, GPT-4o, and Llama-3-8B shows that these models, when making decisions in these games, are affected by at least one of the following systematic biases: positional bias, payoff bias, or behavioural bias. This indicates that LLMs do not fully rely on logical reasoning when making these strategic decisions. As a result, it was found that the LLMs' performance drops when the game configuration is misaligned with the affecting biases. When misaligned, GPT-3.5, GPT-4-Turbo, GPT-4o, and Llama-3-8B show an average performance drop of 32\%, 25\%, 34\%, and 29\% respectively in Stag Hunt, and 28\%, 16\%, 34\%, and 24\% respectively in Prisoner's Dilemma. Surprisingly, GPT-4o (a top-performing LLM across standard benchmarks) suffers the most substantial performance drop, suggesting that newer models are not addressing these issues. Interestingly, we found that a commonly used method of improving the reasoning capabilities of LLMs, chain-of-thought (CoT) prompting, reduces the biases in GPT-3.5, GPT-4o, and Llama-3-8B but increases the effect of the bias in GPT-4-Turbo, indicating that CoT alone cannot fully serve as a robust solution to this problem. We perform several additional experiments, which provide further insight into these observed behaviours.

Bose-Einstein condensates (BECs) have been proposed for many applications in atom interferometry, as their coherence over long evolution times promises unprecedented sensitivity. To date, BECs can be efficiently created in devices using atom chips, but these are still complex and place high demands on size, weight and power. To further simplify these setups, we equipped an atom chip with a nano-structured diffraction-grating to derive all beams for the magneto-optical trap (MOT) from a single laser beam. Moreover, using a 2D$^+$-MOT as an atomic source and a beam with uniform intensity for the grating illumination, we capture $1\times10^9$ atoms in one second, cool them to $14\,\mu$K, and demonstrate magnetic trapping using the atom chip. This is a major step towards the simplification of portable BEC devices for quantum sensing on earth and in space., Comment: 12 pages, 3 figures

Given its real-time capability to quantify mechanical tissue properties, ultrasound shear wave elastography holds significant promise in clinical musculoskeletal imaging. However, existing shear wave elastography methods fall short in enabling full-limb analysis of 3D anatomical structures under diverse loading conditions, and may introduce measurement bias due to sonographer-applied force on the transducer. These limitations pose numerous challenges, particularly for 3D computational biomechanical tissue modeling in areas like prosthetic socket design. In this feasibility study, a clinical linear ultrasound transducer system with integrated shear wave elastography capabilities was utilized to scan both a calibrated phantom and human limbs in a water tank imaging setup. By conducting 2D and 3D scans under varying compressive loads, this study demonstrates the feasibility of volumetric ultrasound shear wave elastography of human limbs. Our preliminary results showcase a potential method for evaluating 3D spatially varying tissue properties, offering future extensions to computational biomechanical modeling of tissue for various clinical scenarios., Comment: Ultrasonic Imaging. 2024

In this work we prove global well-posedness for the massive Maxwell-Dirac system in the Lorenz gauge in $\mathbb{R}^{1+3}$, for small, sufficiently smooth and decaying initial data, as well as modified scattering for the solutions. Heuristically we exploit the close connection between the massive Maxwell-Dirac and the wave-Klein-Gordon equations, while developing a novel approach which applies directly at the level of the Dirac equations. The modified scattering result follows from a precise description of the asymptotic behavior of the solutions inside the light cone, which we derive via the method of testing with wave packets of Ifrim-Tataru., Comment: 64 pages; v2: minor corrections; v3: processing errors concerning the tikz pictures fixed

This paper describes the KnowledgeHub tool, a scientific literature Information Extraction (IE) and Question Answering (QA) pipeline. This is achieved by supporting the ingestion of PDF documents that are converted to text and structured representations. An ontology can then be constructed where a user defines the types of entities and relationships they want to capture. A browser-based annotation tool enables annotating the contents of the PDF documents according to the ontology. Named Entity Recognition (NER) and Relation Classification (RC) models can be trained on the resulting annotations and can be used to annotate the unannotated portion of the documents. A knowledge graph is constructed from these entity and relation triples which can be queried to obtain insights from the data. Furthermore, we integrate a suite of Large Language Models (LLMs) that can be used for QA and summarisation that is grounded in the included documents via a retrieval component. KnowledgeHub is a unique tool that supports annotation, IE and QA, which gives the user full insight into the knowledge discovery pipeline.

Objective: We report the development of the patient-centered myAURA application and suite of methods designed to aid epilepsy patients, caregivers, and researchers in making decisions about care and self-management. Materials and Methods: myAURA rests on the federation of an unprecedented collection of heterogeneous data resources relevant to epilepsy, such as biomedical databases, social media, and electronic health records. A generalizable, open-source methodology was developed to compute a multi-layer knowledge graph linking all this heterogeneous data via the terms of a human-centered biomedical dictionary. Results: The power of the approach is first exemplified in the study of the drug-drug interaction phenomenon. Furthermore, we employ a novel network sparsification methodology using the metric backbone of weighted graphs, which reveals the most important edges for inference, recommendation, and visualization, such as pharmacology factors patients discuss on social media. The network sparsification approach also allows us to extract focused digital cohorts from social media whose discourse is more relevant to epilepsy or other biomedical problems. Finally, we present our patient-centered design and pilot-testing of myAURA, including its user interface, based on focus groups and other stakeholder input. Discussion: The ability to search and explore myAURA's heterogeneous data sources via a sparsified multi-layer knowledge graph, as well as the combination of those layers in a single map, are useful features for integrating relevant information for epilepsy. Conclusion: Our stakeholder-driven, scalable approach to integrate traditional and non-traditional data sources, enables biomedical discovery and data-powered patient self-management in epilepsy, and is generalizable to other chronic conditions.

Scaling deep learning to massive, diverse internet data has yielded remarkably general capabilities in visual and natural language understanding and generation. However, data has remained scarce and challenging to collect in robotics, seeing robot learning struggle to obtain similarly general capabilities. Promising Learning from Videos (LfV) methods aim to address the robotics data bottleneck by augmenting traditional robot data with large-scale internet video data. This video data offers broad foundational information regarding physical behaviour and the underlying physics of the world, and thus can be highly informative for a generalist robot. In this survey, we present a thorough overview of the emerging field of LfV. We outline fundamental concepts, including the benefits and challenges of LfV. We provide a comprehensive review of current methods for extracting knowledge from large-scale internet video, addressing key challenges in LfV, and boosting downstream robot and reinforcement learning via the use of video data. The survey concludes with a critical discussion of challenges and opportunities in LfV. Here, we advocate for scalable foundation model approaches that can leverage the full range of available internet video to improve the learning of robot policies and dynamics models. We hope this survey can inform and catalyse further LfV research, driving progress towards the development of general-purpose robots.

Supercontinua are broadband spectra that are essential to optical spectroscopy, sensing, imaging, and metrology. They are generated from ultrashort laser pulses through nonlinear frequency conversion in fibers, bulk media, and chip-integrated waveguides. For any generating platform, balancing the competing criteria of strong nonlinearity, transparency, and absence of multiphoton absorption is a key challenge. Here, we explore supercontinuum generation in integrated gallium nitride (GaN) waveguides, which combine a high Kerr-nonlinearity, mid-infrared transparency, and a large bandgap that prevents two- and three-photon absorption in the technologically important telecom C-band, where compact erbium-based pump lasers exist. Using this type of laser, we demonstrate tunable dispersive waves and gap-free spectra extending to almost 4 micron in wavelength, relevant to functional group chemical sensing. Additionally, leveraging the material's second-order nonlinearity, we implement on-chip f-2f interferometry to detect the pump laser's carrier-envelope offset frequency, which enables precision metrology. These results demonstrate the versatility of GaN-on-sapphire as a new platform for broadband nonlinear photonics.

Scaling superconducting digital circuits requires fundamental changes in the current material set and fabrication process. The transition to 300 mm wafers and the implementation of advanced lithography are instrumental in facilitating mature CMOS processes, ensuring uniformity, and optimizing the yield. This study explores the properties of NbxTi(1-x)N films fabricated by magnetron DC sputtering on 300 mm Si wafers. As a promising alternative to traditional Nb in device manufacturing, NbxTi(1-x)N offers numerous advantages, including enhanced stability and scalability to smaller dimensions, in both processing and design. As a ternary material, NbxTi(1-x)N allows engineering material parameters by changing deposition conditions. The engineered properties can be used to modulate device parameters through the stack and mitigate failure modes. We report characterization of NbxTi(1-x)N films at less than 2% thickness variability, 2.4% Tc variability and 3% composition variability. The films material properties such as resistivity (140-375 {\Omega}cm) and critical temperature Tc (4.6 K - 14.1 K) are correlated with stoichiometry and morphology of the films. Our results highlight the significant influence of deposition conditions on crystallographic texture along the films and its correlation with Tc., Comment: 8 pages 8 figures

For multi-variable finite measure spaces, we present in this paper a new framework for non-orthogonal $L^2$ Fourier expansions. Our results hold for probability measures $\mu$ with finite support in $\mathbb{R}^d$ that satisfy a certain disintegration condition that we refer to as ``slice-singular''. In this general framework, we present explicit $L^{2}(\mu)$-Fourier expansions, with Fourier exponentials having positive Fourier frequencies in each of the d coordinates. Our Fourier representations apply to every $f \in L^2(\mu)$, are based on an extended Kaczmarz algorithm, and use a new recursive $\mu$ Rokhlin disintegration representation. In detail, our Fourier series expansion for $f$ is in terms of the multivariate Fourier exponentials $\{e_n\}$, but the associated Fourier coefficients for $f$ are now computed from a Kaczmarz system $\{g_n\}$ in $L^{2}(\mu)$ which is dual to the Fourier exponentials. The $\{g_n\}$ system is shown to be a Parseval frame for $L^{2}(\mu)$. Explicit computations for our new Fourier expansions entail a detailed analysis of subspaces of the Hardy space on the polydisk, dual to $L^{2}(\mu)$, and an associated d-variable Normalized Cauchy Transform. Our results extend earlier work for measures $\mu$ in one and two dimensions, i.e., $d=1 (\mu $ singular), and $d=2 (\mu$ assumed slice-singular). Here our focus is the extension to the cases of measures $\mu$ in dimensions $d >2$. Our results are illustrated with the use of explicit iterated function systems (IFSs), including the IFS generated Menger sponge for $d=3$.

Ultracold atomic gases hold unique promise for space science by capitalizing on quantum advantages and extended freefall, afforded in a microgravity environment, to enable next-generation precision sensors. Atom interferometers are a class of quantum sensors which can use freely falling gases of atoms cooled to sub-photon-recoil temperatures to provide unprecedented sensitivities to accelerations, rotations, and gravitational forces, and are currently being developed for space-based applications in gravitational, earth, and planetary sciences, as well as to search for subtle forces that could signify physics beyond General Relativity and the Standard Model. NASA's Cold Atom Lab (CAL) operates onboard the International Space Station as a multi-user facility for studies of ultracold atoms and to mature quantum technologies, including atom interferometry, in persistent microgravity. In this paper, we report on path-finding experiments utilizing ultracold $^{87}$Rb atoms in the CAL atom interferometer, which was enabled by an on-orbit upgrade of the CAL science module: A three-pulse Mach-Zehnder interferometer was studied to understand limitations from the influence of ISS vibrations. Additionally, Ramsey shear-wave interferometry was used to manifest interference patterns in a single run that were observable for over 150 ms free-expansion time. Finally, the CAL atom interferometer was used to remotely measure the photon recoil from the atom interferometer laser as a demonstration of the first quantum sensor using matter-wave interferometry in space., Comment: 14 pages, 5 figures

The adaption of multilingual pre-trained LLMs into eloquent and helpful assistants is essential to facilitate their use across different language regions. In that spirit, we are the first to conduct an extensive study of the performance of multilingual models instruction-tuned on different language compositions on parallel instruction-tuning benchmarks across a selection of the most spoken Indo-European languages. We systematically examine the effects of language and instruction dataset size on a mid-sized and a large, multilingual LLMs by instruction-tuning them on parallel instruction-tuning datasets. Our results demonstrate that instruction-tuning on parallel instead of monolingual corpora benefits cross-lingual instruction following capabilities by up to 9.9%. Furthermore, we show that the Superficial Alignment Hypothesis does not hold in general, as the investigated multilingual 7B parameter model presents a counter-example requiring large-scale instruction-tuning datasets. Finally, we conduct a human annotation study to understand the alignment between human-based and GPT-4-based evaluation within multilingual chat scenarios., Comment: Accepted for EMNLP 2024 (Main), 27 pages, 8 figures

Microresonator frequency combs (microcombs) hold great potential for precision metrology within a compact form factor, impacting a wide range of applications such as point-of-care diagnostics, environmental monitoring, time-keeping, navigation and astronomy. Through the principle of self-injection locking, electrically-driven chip-based microcombs with minimal complexity are now feasible. However, phase-stabilisation of such self-injection-locked microcombs, a prerequisite for metrological frequency comb, has not yet been attained. Here, we address this critical need by demonstrating full phase-stabilisation of a self-injection-locked microcomb. The microresonator is implemented in a silicon nitride photonic chip, and by controlling a pump laser diode and a microheater with low voltage signals (less than 1.5 V), we achieve independent control of the comb's offset and repetition rate frequencies. Both actuators reach a bandwidth of over 100 kHz, enabling phase-locking of the microcomb to external frequency references. These results establish photonic chip-based, self-injection-locked microcombs as low-complexity yet versatile sources for coherent precision metrology in emerging applications., Comment: 12 pages, 7 figures

The development of quantum computing systems for large scale algorithms requires targeted error rates unachievable through hardware advancements alone. Quantum Error Correction (QEC) allows us to use systems with a large number of physical qubits to form a fault tolerant system with a lower number of logical qubits and a favourable logical error rate. While some gates can be easily implemented in a QEC code transversally, there is no code that has a universal set of transversal gates. Transversal Injection is a new method of preparing logical non-Pauli eigenstates that can be used as resource states for quantum computation. State preparation can be done directly in the surface code and has the potential to prepare higher fidelity injected states. Compared to other techniques, transversal injection can reduce the resource burden for state distillation protocols. In this paper, the authors present the theory behind this new technique as well as an algorithm for calculating the resulting logical states prepared in the surface code., Comment: 6 Pages, 3 Figures, IEEE QEC23. arXiv admin note: substantial text overlap with arXiv:2211.10046

Femtosecond laser pulses enable the synthesis of light across the electromagnetic spectrum and provide access to ultrafast phenomena in physics, biology, and chemistry. Chip-integration of femtosecond technology could revolutionize applications such as point-of-care diagnostics, bio-medical imaging, portable chemical sensing, or autonomous navigation. However, current chip-integrated pulse sources lack the required peak power and on-chip amplification of femtosecond pulses has been an unresolved challenge. Here, addressing this challenge, we report >50-fold amplification of 1 GHz-repetition-rate chirped femtosecond pulses in a CMOS-compatible photonic chip to 800 W peak power with 116 fs pulse duration. This power level is 2-3 orders of magnitude higher compared to those in previously demonstrated on-chip pulse sources and can provide the power needed to address key applications. To achieve this, detrimental nonlinear effects are mitigated through all-normal dispersion, large mode-area and rare-earth-doped gain waveguides. These results offer a pathway to chip-integrated femtosecond technology with peak power-levels characteristic of table-top sources., Comment: 15 pages, 7 figures

Photonics offer numerous functionalities that can be used to realize astrophotonic instruments. The most spectacular example to date is the ESO Gravity instrument at the Very Large Telescope in Chile. Integrated astrophotonic devices stand to offer critical advantages for instrument development, including extreme miniaturization, as well as integration, superior thermal and mechanical stabilization owing to the small footprint, and high replicability offering cost savings. Numerous astrophotonic technologies have been developed to address shortcomings of conventional instruments to date, including for example the development of photonic lanterns, complex aperiodic fiber Bragg gratings, complex beam combiners to enable long baseline interferometry, and laser frequency combs for high precision spectral calibration of spectrometers. Despite these successes, the facility implementation of photonic solutions in astronomical instrumentation is currently limited because of (1) low throughputs from coupling to fibers, coupling fibers to chips, propagation and bend losses, device losses, etc, (2) difficulties with scaling to large channel count devices needed for large bandwidths and high resolutions, and (3) efficient integration of photonics with detectors, to name a few. In this roadmap, we identify 24 areas that need further development. We outline the challenges and advances needed across those areas covering design tools, simulation capabilities, fabrication processes, the need for entirely new components, integration and hybridization and the characterization of devices. To realize these advances the astrophotonics community will have to work cooperatively with industrial partners who have more advanced manufacturing capabilities. With the advances described herein, multi-functional instruments will be realized leading to novel observing capabilities for both ground and space platforms., Comment: 191 pages, 47 figures. This is the version of the article before peer review or editing, as submitted by an author to J. Phys. Photonics. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at https://iopscience.iop.org/article/10.1088/2515-7647/ace869/meta

Frequency combs from continuous-wave-driven Kerr-nonlinear microresonators have evolved into a key photonic technology with applications from optical communication to precision spectroscopy. Essential to many of these applications is the control of the comb's defining parameters, i.e., carrier-envelope offset frequency and repetition rate. An elegant and all-optical approach to controlling both degrees of freedom is the suitable injection of a secondary continuous-wave laser into the resonator onto which one of the comb lines locks. Here, we study experimentally such sideband injection locking in microresonator soliton combs across a wide optical bandwidth and derive analytic scaling laws for the locking range and repetition rate control. As an application example, we demonstrate optical frequency division and repetition rate phase-noise reduction to three orders of magnitude below the noise of a free-running system. The presented results can guide the design of sideband injection-locked, parametrically generated frequency combs with opportunities for low-noise microwave generation, compact optical clocks with simplified locking schemes and more generally, all-optically stabilized frequency combs from Kerr-nonlinear resonators., Comment: 13 pages, 6 figures

The recent success of Large Language Models (LLMs) has been predominantly driven by curating the training dataset composition, scaling of model architectures and dataset sizes and advancements in pretraining objectives, leaving tokenizer influence as a blind spot. Shedding light on this underexplored area, we conduct a comprehensive study on the influence of tokenizer choice on LLM downstream performance by training 24 mono- and multilingual LLMs at a 2.6B parameter scale, ablating different tokenizer algorithms and parameterizations. Our studies highlight that the tokenizer choice can significantly impact the model's downstream performance and training costs. In particular, we find that the common tokenizer evaluation metrics fertility and parity are not always predictive of model downstream performance, rendering these metrics a questionable proxy for the model's downstream performance. Furthermore, we show that multilingual tokenizers trained on the five most frequent European languages require vocabulary size increases of factor three in comparison to English. While English-centric tokenizers have been applied to the training of multi-lingual LLMs in the past, we find that this approach results in a severe downstream performance degradation and additional training costs of up to 68%, due to an inefficient tokenization vocabulary.