Search

When a chemical bond is broken, the molecular structure undergoes a transformation. An ideal experiment should probe the change in the electronic and nuclear structure simultaneously. Here, we present a method for the simultaneous time-resolved imaging of nuclear and electron dynamics by combining Coulomb explosion imaging with strong-field photoelectron momentum imaging. The simplest chemical reaction, H$_2^+$ $\rightarrow$ H$^+ +$ H, is probed experimentally for the delay-dependent kinetic energy release, and numerically for the transient change in the photoelectron spectra during the dissociation process. The three-dimensional Schr\"odinger equation is solved in the fixed-nuclei approximation numerically and the results are compared to those from a simple imaging model. The numerical results reflect the evolution in the electron density in the molecular ion as its bond is first stretched and then brakes apart. Our work shows how simple gas-phase chemical dynamics can be captured in complete molecular movies.

By framing reinforcement learning as a sequence modeling problem, recent work has enabled the use of generative models, such as diffusion models, for planning. While these models are effective in predicting long-horizon state trajectories in deterministic environments, they face challenges in dynamic settings with moving obstacles. Effective collision avoidance demands continuous monitoring and adaptive decision-making. While replanning at every timestep could ensure safety, it introduces substantial computational overhead due to the repetitive prediction of overlapping state sequences -- a process that is particularly costly with diffusion models, known for their intensive iterative sampling procedure. We propose an adaptive generative planning approach that dynamically adjusts replanning frequency based on the uncertainty of action predictions. Our method minimizes the need for frequent, computationally expensive, and redundant replanning while maintaining robust collision avoidance performance. In experiments, we obtain a 13.5% increase in the mean trajectory length and a 12.7% increase in mean reward over long-horizon planning, indicating a reduction in collision rates and an improved ability to navigate the environment safely.

Time-series forecasting has seen significant advancements with the introduction of token prediction mechanisms such as multi-head attention. However, these methods often struggle to achieve the same performance as in language modeling, primarily due to the quadratic computational cost and the complexity of capturing long-range dependencies in time-series data. State-space models (SSMs), such as Mamba, have shown promise in addressing these challenges by offering efficient solutions with linear RNNs capable of modeling long sequences with larger context windows. However, there remains room for improvement in accuracy and scalability. We propose the use of Test-Time Training (TTT) modules in a parallel architecture to enhance performance in long-term time series forecasting. Through extensive experiments on standard benchmark datasets, we demonstrate that TTT modules consistently outperform state-of-the-art models, including the Mamba-based TimeMachine, particularly in scenarios involving extended sequence and prediction lengths. Our results show significant improvements in Mean Squared Error (MSE) and Mean Absolute Error (MAE), especially on larger datasets such as Electricity, Traffic, and Weather, underscoring the effectiveness of TTT in capturing long-range dependencies. Additionally, we explore various convolutional architectures within the TTT framework, showing that even simple configurations like 1D convolution with small filters can achieve competitive results. This work sets a new benchmark for time-series forecasting and lays the groundwork for future research in scalable, high-performance forecasting models.

This paper explores the potential of a small, domain-specific language model trained exclusively on sports-related data. We investigate whether extensive training data with specially designed small model structures can overcome model size constraints. The study introduces the OnlySports collection, comprising OnlySportsLM, OnlySports Dataset, and OnlySports Benchmark. Our approach involves: 1) creating a massive 600 billion tokens OnlySports Dataset from FineWeb, 2) optimizing the RWKV architecture for sports-related tasks, resulting in a 196M parameters model with 20-layer, 640-dimension structure, 3) training the OnlySportsLM on part of OnlySports Dataset, and 4) testing the resultant model on OnlySports Benchmark. OnlySportsLM achieves a 37.62%/34.08% accuracy improvement over previous 135M/360M state-of-the-art models and matches the performance of larger models such as SomlLM 1.7B and Qwen 1.5B in the sports domain. Additionally, the OnlySports collection presents a comprehensive workflow for building high-quality, domain-specific language models, providing a replicable blueprint for efficient AI development across various specialized fields., Comment: 13 pages, 4 figures, 4 tables

As humankind prepares to establish outposts and infrastructure on the Moon, the ability to manufacture parts and buildings on-site is crucial. While transporting raw materials from Earth can be costly and time-consuming, in-situ resource utilization (ISRU) presents an attractive alternative. This review paper aims to provide a thorough examination of the current state and future potential of Lunar-based manufacturing and construction (LBMC), with a particular focus on the prospect of utilizing in-situ resources and additive manufacturing. The paper analyzes existing research on LBMC from various perspectives, including different manufacturing techniques and compositions, the potential of ISRU for LBMC, characterization of built parts and structures, the role of energy sources and efficiency, the impact of low-gravity and vacuum conditions, and the feasibility of using artificial intelligence, automation, and robotics. By synthesizing these findings, this review offers valuable insights into the challenges and opportunities that lie ahead for LBMC.

Planning with generative models has emerged as an effective decision-making paradigm across a wide range of domains, including reinforcement learning and autonomous navigation. While continuous replanning at each timestep might seem intuitive because it allows decisions to be made based on the most recent environmental observations, it results in substantial computational challenges, primarily due to the complexity of the generative model's underlying deep learning architecture. Our work addresses this challenge by introducing a simple adaptive planning policy that leverages the generative model's ability to predict long-horizon state trajectories, enabling the execution of multiple actions consecutively without the need for immediate replanning. We propose to use the predictive uncertainty derived from a Deep Ensemble of inverse dynamics models to dynamically adjust the intervals between planning sessions. In our experiments conducted on locomotion tasks within the OpenAI Gym framework, we demonstrate that our adaptive planning policy allows for a reduction in replanning frequency to only about 10% of the steps without compromising the performance. Our results underscore the potential of generative modeling as an efficient and effective tool for decision-making.

The potentials of Generative-AI technologies like Large Language models (LLMs) to revolutionize education are undermined by ethical considerations around their misuse which worsens the problem of academic dishonesty. LLMs like GPT-4 and Llama 2 are becoming increasingly powerful in generating sophisticated content and answering questions, from writing academic essays to solving complex math problems. Students are relying on these LLMs to complete their assignments and thus compromising academic integrity. Solutions to detect LLM-generated text are compute-intensive and often lack generalization. This paper presents a novel approach for detecting LLM-generated AI-text using a visual representation of word embedding. We have formulated a novel Convolutional Neural Network called ZigZag ResNet, as well as a scheduler for improving generalization, named ZigZag Scheduler. Through extensive evaluation using datasets of text generated by six different state-of-the-art LLMs, our model demonstrates strong intra-domain and inter-domain generalization capabilities. Our best model detects AI-generated text with an impressive average detection rate (over inter- and intra-domain test data) of 88.35%. Through an exhaustive ablation study, our ZigZag ResNet and ZigZag Scheduler provide a performance improvement of nearly 4% over the vanilla ResNet. The end-to-end inference latency of our model is below 2.5ms per sentence. Our solution offers a lightweight, computationally efficient, and faster alternative to existing tools for AI-generated text detection, with better generalization performance. It can help academic institutions in their fight against the misuse of LLMs in academic settings. Through this work, we aim to contribute to safeguarding the principles of academic integrity and ensuring the trustworthiness of student work in the era of advanced LLMs., Comment: 11 pages, 5 figures

This article seeks to contribute to a nuanced understanding of the integration of Design Structure Matrix (DSM) and genetic algorithms in the context of Complex Systems modelling described within Model-Based System Engineering approach. By examining coupling minimization as a critical aspect of advanced systems engineering practices, we aim to provide a scholarly exploration, blending theoretical insights with practical applications. The objective is to equip systems architects with analytical tools integrated within their Model Based Systems Engineering (MBSE) environment for exploring the design space of component interactions, facilitating the identification of optimal system architectures., Comment: ERTS2024, SEE; 3AF, Jun 2024, Toulouse, France

The aim of this paper is to investigate further properties of $z$-elements in multiplicative lattices. We utilize $z$-closure operators to extend several properties of $z$-ideals to $z$-elements and introduce various distinguished subclasses of $z$-elements, such as $z$-prime, $z$-semiprime, $z$-primary, $z$-irreducible, and $z$-strongly irreducible elements, and study their properties. We provide a characterization of multiplicative lattices where $z$-elements are closed under finite products and a representation of $z$-elements in terms of $z$-irreducible elements in $z$-Noetherian multiplicative lattices., Comment: 15 pages

The objective of this paper is to extend certain properties observed in $d$-ideals of rings and $d$-elements of frames to Baer elements in multiplicative lattices introduced in D. D. Anderson, C. Jayaram, and P. A. Phiri, Baer lattices, \textit{Acta Sci. Math. (Szeged)}, 59 (1994), 61--74. Additionally, we present results concerning these elements that have not been addressed in the study of $d$-ideals of rings. Furthermore, we introduce Baer closures and explore Baer maximal, prime, semiprime, and meet-irreducible elements., Comment: 12 apges

Paraxial diffraction modeling based on the Fourier transform has seen widespread implementation for simulating the response of a diffraction-limited optical system. For systems where the paraxial assumption is not sufficient, a class of algorithms has been developed that employs hybrid propagation physics to compute the propagation of an elementary beamlet along geometric ray paths. These "beamlet decomposition" algorithms include the well-known Gaussian Beamlet Decomposition (GBD) algorithm, of which several variants have been created. To increase the computational efficiency of the GBD algorithm, we derive an alternative expression of the technique that utilizes the analytical propagation of beamlets to tilted planes. We then use this accelerated algorithm to conduct a parameter-space search to find the optimal combination of free parameters in GBD to construct the analytical Airy function. The experiment is conducted on a consumer-grade CPU, and a high-performance GPU, where the new algorithm is 34 times faster than the previously published algorithm on CPUs, and 67,513 times faster on GPUs., Comment: 19 pages, 10 figures, Accepted to Optics Express on April 18 2024

This paper focuses on the design and systematic evaluation of fabric-based, bellow-type soft pneumatic actuators to assist with flexion and extension of the elbow, intended for use in infant wearable devices. Initially, the performance of a range of actuator variants was explored via simulation. The actuator variants were parameterized based on the shape, number, and size of the cells present. Subsequently, viable actuator variants identified from the simulations were fabricated and underwent further testing on a physical model based on an infant's body anthropometrics. The performance of these variants was evaluated based on kinematic analyses using metrics including movement smoothness, path length, and elbow joint angle. Internal pressure of the actuators was also attained. Taken together, results reported herein provide valuable insights about the suitability of several actuator designs to serve as components for pediatric wearable assistive devices.

This paper proposes MapGPT which is a novel approach that integrates the capabilities of language models, specifically large language models (LLMs), with spatial data processing techniques. This paper introduces MapGPT, which aims to bridge the gap between natural language understanding and spatial data analysis by highlighting the relevant core building blocks. By combining the strengths of LLMs and geospatial analysis, MapGPT enables more accurate and contextually aware responses to location-based queries. The proposed methodology highlights building LLMs on spatial and textual data, utilizing tokenization and vector representations specific to spatial information. The paper also explores the challenges associated with generating spatial vector representations. Furthermore, the study discusses the potential of computational capabilities within MapGPT, allowing users to perform geospatial computations and obtain visualized outputs. Overall, this research paper presents the building blocks and methodology of MapGPT, highlighting its potential to enhance spatial data understanding and generation in natural language processing applications.

The Roman Space Telescope will have the first advanced coronagraph in space, with deformable mirrors for wavefront control, low-order wavefront sensing and maintenance, and a photon-counting detector. It is expected to be able to detect and characterize mature, giant exoplanets in reflected visible light. Over the past decade the performance of the coronagraph in its flight environment has been simulated with increasingly detailed diffraction and structural/thermal finite element modeling. With the instrument now being integrated in preparation for launch within the next few years, the present state of the end-to-end modeling is described, including the measured flight components such as deformable mirrors. The coronagraphic modes are thoroughly described, including characteristics most readily derived from modeling. The methods for diffraction propagation, wavefront control, and structural and thermal finite-element modeling are detailed. The techniques and procedures developed for the instrument will serve as a foundation for future coronagraphic missions such as the Habitable Worlds Observatory., Comment: 113 pages, 85 figures, to be published in SPIE Journal of Astronomical Telescopes, Instruments, and Systems

Integrated optical models allow for accurate prediction of the as-built performance of an optical instrument. Optical models are typically composed of a separate ray trace and diffraction model to capture both the geometrical and physical regimes of light. These models are typically separated across both open-source and commercial software that don't interface with each other directly. To bridge the gap between ray trace models and diffraction models, we have built an open-source optical analysis platform in Python called Poke that uses commercial ray tracing APIs and open-source physical optics engines to simultaneously model scalar wavefront error, diffraction, and polarization. Poke operates by storing ray data from a commercial ray tracing engine into a Python object, from which physical optics calculations can be made. We present an introduction to using Poke, and highlight the capabilities of two new propagation physics modules that add to the utility of existing scalar diffraction models. Gaussian Beamlet Decomposition is a ray-based approach to diffraction modeling that allows us to integrate physical optics models with ray trace models to directly capture the influence of ray aberrations in diffraction simulations. Polarization Ray Tracing is a ray-based method of vector field propagation that can diagnose the polarization aberrations in optical systems. Poke has been recently used to study the next generation of astronomical observatories, including the ground-based Extremely Large Telescopes and a 6 meter space telescope early concept for NASA's Habitable Worlds Observatory., Comment: 11 Pages, 9 Figures, Published in Proceedings of SPIE Optical Modeling and Performance Predictions XIII Paper 12664-5

Accurate predictions of electricity demands are necessary for managing operations in a small aggregation load setting like a Microgrid. Due to low aggregation, the electricity demands can be highly stochastic and point estimates would lead to inflated errors. Interval estimation in this scenario, would provide a range of values within which the future values might lie and helps quantify the errors around the point estimates. This paper introduces a residual bootstrap algorithm to generate interval estimates of day-ahead electricity demand. A machine learning algorithm is used to obtain the point estimates of electricity demand and respective residuals on the training set. The obtained residuals are stored in memory and the memory is further partitioned. Days with similar demand patterns are grouped in clusters using an unsupervised learning algorithm and these clusters are used to partition the memory. The point estimates for test day are used to find the closest cluster of similar days and the residuals are bootstrapped from the chosen cluster. This algorithm is evaluated on the real electricity demand data from EULR(End Use Load Research) and is compared to other bootstrapping methods for varying confidence intervals.

Affect is an emotional characteristic encompassing valence, arousal, and intensity, and is a crucial attribute for enabling authentic conversations. While existing text-to-speech (TTS) and speech-to-speech systems rely on strength embedding vectors and global style tokens to capture emotions, these models represent emotions as a component of style or represent them in discrete categories. We propose AffectEcho, an emotion translation model, that uses a Vector Quantized codebook to model emotions within a quantized space featuring five levels of affect intensity to capture complex nuances and subtle differences in the same emotion. The quantized emotional embeddings are implicitly derived from spoken speech samples, eliminating the need for one-hot vectors or explicit strength embeddings. Experimental results demonstrate the effectiveness of our approach in controlling the emotions of generated speech while preserving identity, style, and emotional cadence unique to each speaker. We showcase the language-independent emotion modeling capability of the quantized emotional embeddings learned from a bilingual (English and Chinese) speech corpus with an emotion transfer task from a reference speech to a target speech. We achieve state-of-art results on both qualitative and quantitative metrics.

Floods are one of the most common natural disasters, with a disproportionate impact in developing countries that often lack dense streamflow gauge networks. Accurate and timely warnings are critical for mitigating flood risks, but hydrological simulation models typically must be calibrated to long data records in each watershed. Using AI, we achieve reliability in predicting extreme riverine events in ungauged watersheds at up to a 5-day lead time that is similar to or better than the reliability of nowcasts (0-day lead time) from a current state of the art global modeling system (the Copernicus Emergency Management Service Global Flood Awareness System). Additionally, we achieve accuracies over 5-year return period events that are similar to or better than current accuracies over 1-year return period events. This means that AI can provide flood warnings earlier and over larger and more impactful events in ungauged basins. The model developed in this paper was incorporated into an operational early warning system that produces publicly available (free and open) forecasts in real time in over 80 countries. This work highlights a need for increasing the availability of hydrological data to continue to improve global access to reliable flood warnings.

This paper outlines the roadmap towards the redefinition of the second, which was recently updated by the CCTF Task Force created by the CCTF in 2020. The main achievements and the open challenges related to the status of the optical frequency standards, their contribution to time scales and UTC, the possibility of their comparison and the knowledge of the Earth's gravitational potential at the necessary level of uncertainty are discussed. In addition, the mandatory criteria to be achieved before redefinition and their current fulfilment level, together with the redefinition options based on a single or on a set of transitions are described., Comment: 26 pages. This paper is based on the work of a CCTF Task Force on the roadmap to the redefinition of the second

The increasing reliance on large language models (LLMs) in academic writing has led to a rise in plagiarism. Existing AI-generated text classifiers have limited accuracy and often produce false positives. We propose a novel approach using natural language processing (NLP) techniques, offering quantifiable metrics at both sentence and document levels for easier interpretation by human evaluators. Our method employs a multi-faceted approach, generating multiple paraphrased versions of a given question and inputting them into the LLM to generate answers. By using a contrastive loss function based on cosine similarity, we match generated sentences with those from the student's response. Our approach achieves up to 94% accuracy in classifying human and AI text, providing a robust and adaptable solution for plagiarism detection in academic settings. This method improves with LLM advancements, reducing the need for new model training or reconfiguration, and offers a more transparent way of evaluating and detecting AI-generated text., Comment: 10 Pages, 4 Figures, 9 Tables, to be published in 18th Workshop on Innovative Use of NLP for Building Educational Applications

The aim of this paper is to study some distinguished classes of $k$-ideals of semirings, which include $k$-prime, $k$-semiprime, $k$-radical, $k$-irreducible, and $k$-strongly irreducible ideals. We discuss some of the properties of $k$-ideals and their products, intersections, and ideal quotients under semiring homomorphisms.