Search Results (7)

Intelligent mining risk assessment (MIRA) is a vital approach for enhancing safety and operational efficiency in mining. In this study, we introduce MIRA-ChatGLM, which leverages pre-trained large language models (LLMs) for the domain of gas risk assessment in coal mines. We meticulously constructed a dataset specifically designed for mining risk analysis and performed parameter-efficient fine-tuning on the locally deployed GLM-4-9B-chat base model to develop MIRA-ChatGLM. By utilizing consumer-grade GPUs and employing LoRA and various levels of quantization algorithms such as QLoRA, we investigated the impact of different data scales and instruction settings on model performance. The evaluation results show that MIRA-ChatGLM achieved excellent performance with BLEU-4, ROUGE-1, ROUGE-2, and ROUGE-L scores of 84.47, 90.63, 86.88, and 90.63, respectively, highlighting its outstanding performance in coal mine gas risk assessment. Through comparative experiments with other large language models of similar size and manual evaluation, MIRA-ChatGLM demonstrated superior performance across multiple key metrics, fully demonstrating its tremendous potential in intelligent mine risk assessment and decision support. Full article

Generating accurate and contextually rich captions for images and videos is essential for various applications, from assistive technology to content recommendation. However, challenges such as maintaining temporal coherence in videos, reducing noise in large-scale datasets, and enabling real-time captioning remain significant. We introduce MIRA-CAP (Memory-Integrated Retrieval-Augmented Captioning), a novel framework designed to address these issues through three core innovations: a cross-modal memory bank, adaptive dataset pruning, and a streaming decoder. The cross-modal memory bank retrieves relevant context from prior frames, enhancing temporal consistency and narrative flow. The adaptive pruning mechanism filters noisy data, which improves alignment and generalization. The streaming decoder allows for real-time captioning by generating captions incrementally, without requiring access to the full video sequence. Evaluated across standard datasets like MS COCO, YouCook2, ActivityNet, and Flickr30k, MIRA-CAP achieves state-of-the-art results, with high scores on CIDEr, SPICE, and Polos metrics, underscoring its alignment with human judgment and its effectiveness in handling complex visual and temporal structures. This work demonstrates that MIRA-CAP offers a robust, scalable solution for both static and dynamic captioning tasks, advancing the capabilities of vision–language models in real-world applications. Full article

The conditions in Mira variable atmospheres make them wonderful laboratories to study a variety of stellar physics such as molecule–grain formation, dust production, shock chemistry, stellar winds, mass loss, opacity-driven pulsation, and shocks. We were awarded an NSF grant to analyze over a decade of synoptic observations from the Palomar Testbed Interferometer (PTI) of 106 Miras to curate a Mira Reference Dataset. The Miras included in this dataset include M-types, S-types, and C-types, and span a wide range of pulsation periods. PTI measured K-band angular sizes that when combined with a distance allow us to directly determine fundamental stellar parameters such as effective temperature, radial size, and bolometric flux. Supplementing observations with interferometric measurements of the stars opens the Mira laboratory to a wealth of different experiments. We provide two case studies to serve as examples of the power of the Mira Reference Dataset. The first case study describes combining PTI measurements with Spitzer IRS spectra of M-type Miras, which allowed us to fully characterize ${\mathrm{CO}}_2$ gas in their atmospheres. The second case study examines how PTI narrow-band data can be used to study phase-dependent pulsation effects on the stellar atmosphere. We provide a list of all the Miras (with coordinates) included in the set for anyone who would like to add them to their observing programs. All the data we produce and collate for this Mira Reference Dataset will be hosted and curated on a website open to the public so that other researchers and citizen scientists can participate in expanding the utility and body of knowledge on this set of “wonderful” stars. Full article

In this paper, we present an analysis of the effectiveness of various machine learning algorithms in classifying astronomical objects using data from the third release (DR3) of the Gaia space mission. The dataset used includes spectral information from the satellite’s red and blue spectrophotometers. The primary goal is to achieve reliable classification with high confidence for symbiotic stars, planetary nebulae, and red giants. Symbiotic stars are binary systems formed by a high-temperature star (a white dwarf in most cases) and an evolved star (Mira type or red giant star); their spectra varies between the typical for these objects (depending on the orbital phase of the object) and present emission lines similar to those observed in PN spectra, which is the reason for this first selection. Several classification algorithms are evaluated, including Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Networks (ANN), Gradient Boosting (GB), and Naive Bayes classifier. The evaluation is based on different metrics such as Precision, Recall, F1-Score, and the Kappa index. The study confirms the effectiveness of classifying the mentioned stars using only their spectral information. The models trained with Artificial Neural Networks and Random Forest demonstrated superior performance, surpassing an accuracy rate of 94.67%. Full article

We processed five years of measurements (2018–2022) of a vertically pointing radar MIRA 35c at the Milešovka meteorological observatory with the aim of analyzing the cloud structure of thunderstorms and comparing differences in measured data for cases when lightning discharges were observed very close to the radar position, and for cases when lightning discharges were observed at a greater distance from the radar position. The MIRA 35c radar is a Doppler polarimetric radar working at 35 GHz (Ka-band) with a vertical resolution of 28.9 m and a time resolution of approximately 2 s. For the analysis, we considered radar data whose radar reflectivity was at least 10 dBZ at 5 km or higher above the radar to ensure that there was a cloud above the radar. We divided the radar data into “near” data (a lightning discharge was registered up to 1 km from the radar position) and “far” data (a lightning discharge was registered from 7.5 to 10 km from the radar position). We compared the following quantities: (i) Power in co-channel (pow), (ii) power in cross-channel (pow-cx), (iii) phase in co-channel (pha), (iv) phase in cross-channel (pha-cx), (v) equivalent radar reflectivity (Ze), (vi) Linear Depolarization Ratio (LDR), (vii) co-polar correlation coefficient (RHO), (viii) Doppler radial velocity (V), (ix) Doppler spectrum width (RMS), and (x) Differential phase (Phi). Pow, pow-cx, pha, pha-cx, and V are basic data measured by the radar, while Ze, LDR, RHO, RMS, and Phi are derived quantities. Our results showed that the characteristics of the compared radar quantities are clearly distinct for “near” dataset from “far” dataset. Furthermore, we found out that there is a clear evolution close to the time of discharges of the observed radar quantities in the “near” dataset, which is not that obvious in the “far” dataset. Full article

During the last two years, the emergence of SARS-CoV-2 has led to millions of deaths worldwide, with a devastating socio-economic impact on a global scale. The scientific community’s focus has recently shifted towards the association of the T cell immunological repertoire with COVID-19 progression and severity, by utilising T cell receptor sequencing (TCR-Seq) assays. The Multiplexed Identification of T cell Receptor Antigen (MIRA) dataset, which is a subset of the immunoACCESS study, provides thousands of TCRs that can specifically recognise SARS-CoV-2 epitopes. Our study proposes a novel Machine Learning (ML)-assisted approach for analysing TCR-Seq data from the antigens’ point of view, with the ability to unveil key antigens that can accurately distinguish between MIRA COVID-19-convalescent and healthy individuals based on differences in the triggered immune response. Some SARS-CoV-2 antigens were found to exhibit equal levels of recognition by MIRA TCRs in both convalescent and healthy cohorts, leading to the assumption of putative cross-reactivity between SARS-CoV-2 and other infectious agents. This hypothesis was tested by combining MIRA with other public TCR profiling repositories that host assays and sequencing data concerning a plethora of pathogens. Our study provides evidence regarding putative cross-reactivity between SARS-CoV-2 and a wide spectrum of pathogens and diseases, with M. tuberculosis and Influenza virus exhibiting the highest levels of cross-reactivity. These results can potentially shift the emphasis of immunological studies towards an increased application of TCR profiling assays that have the potential to uncover key mechanisms of cell-mediated immune response against pathogens and diseases. Full article

This paper describes an automatic multi-image robust alignment (MIRA) procedure able to simultaneously co-register a time series of medium-resolution satellite images in a bundle block adjustment (BBA) fashion. Instead of the direct co-registration of each image with respect to a reference ‘master’ image on the basis of corresponding features, MIRA also considers those tie points that may be not be shared with the master, but they only connect the other images (‘slaves’) among them. In a first stage, tie points are automatically extracted by using pairwise feature-based matching based on the SURF operator. In a second stage, such extracted features are re-ordered to find corresponding tie points visible on multiple image pairs. A ‘master’ image is then selected with the only purpose to establish the datum of the final image alignment and to instantiate the computation of approximate registration parameters. All the available information obtained so far is fed into a least-squares BBA to estimate the unknowns, which include the registration parameters and the coordinates of tie points re-projected in the ‘master’ image space. The analysis of inner and outer reliability of the observations is applied to assess whether the residual blunders may be located using data snooping, and to evaluate the influence of undetected outliers on the final registration results. Three experiments with simulated datasets and one example consisting of eleven Landsat-5/TM images are reported and discussed. In the case of real data, results have been positively checked against the ones obtained by using alternative procedures (BBA with manual measurements and ‘slave-to-master’ registration based on automatically extracted tie points). These experiments have confirmed the correctness of the MIRA approach and have highlighted the potential of the inner control on the final quality of the solution that may come from the reliability analysis. Full article