Satellite video tracking presents significant challenges due to unpredictable target variations, environmental disturbances, and occlusions. Existing approaches either rely on auxiliary modalities or require full fine-tuning of foundation models, resulting in excessive parameter sensitivity and poor generalization. Meanwhile, conventional prompt-based tuning only updates parameters at a single location, limiting its ability to adapt to complex appearance changes. To address these limitations, we propose Adaptive Visual Prompting for Effective Satellite Video Tracking (AVPTrack). Unlike conventional prompts, introduced Super Prompts dynamically refine the original template at multiple distinct positions. This multi-location adaptation allows for fine-grained representation learning, enabling the tracker to better capture target variations and resist environmental disturbances. Additionally, Dynamic Templates are introduced to mitigate tracking failures in highly challenging scenarios, such as occlusions and background clutter, ensuring robust target localization. Furthermore, the Template Selection Adapter (TSA) selects the most relevant templates in real-time, enhancing tracking efficiency. These components are optimized during training while keeping other parameters frozen, ensuring parameter efficiency. We also investigate the relationship between fine-tuning proportions and learning rates to optimize model performance. Extensive evaluations on the SV248S, SatSOT, and VISO datasets demonstrate the superior adaptability and robustness of AVPTrack compared to existing methods.

Intellectual Property (IP) management involves strategically protecting and utilizing intellectual assets to enhance organizational innovation, competitiveness, and value creation. Patent matching is a crucial task in intellectual property management, which facilitates the organization and utilization of patents. Existing models often rely on the emergent capabilities of Large Language Models (LLMs) and leverage them to identify related patents directly. However, these methods usually depend on matching keywords and overlook the hierarchical classification and categorical relationships of patents. In this paper, we propose MemGraph, a method that augments the patent matching capabilities of LLMs by incorporating a memory graph derived from their parametric memory. Specifically, MemGraph prompts LLMs to traverse their memory to identify relevant entities within patents, followed by attributing these entities to corresponding ontologies. After traversing the memory graph, we utilize extracted entities and ontologies to improve the capability of LLM in comprehending the semantics of patents. Experimental results on the PatentMatch dataset demonstrate the effectiveness of MemGraph, achieving a 17.68% performance improvement over baseline LLMs. The further analysis highlights the generalization ability of MemGraph across various LLMs, both in-domain and out-of-domain, and its capacity to enhance the internal reasoning processes of LLMs during patent matching. All data and codes are available at https://github.com/NEUIR/MemGraph.

Fetal electrocardiography (FECG) presents an important avenue for continuous fetal monitoring. Nonetheless, effectively extracting FECG from maternal electrocardiogram (MECG) is one considerable challenge due to its weaker amplitude compared to MECG and the non-Gaussian nature of background noise. In this letter, we introduce alpha-stable noise to model the realistic interference due to its high scalability. To improve the accuracy of FECG extraction under impulsive noise (alpha-stable noise with strong impulses), we introduce the least Lncosh algorithm (Llncosh) and the improved Llncosh algorithm (ILL) is proposed based on the Amplitude Hyperbolic Tangent Transformation (AHTT) to optimize the preset parameter. Moreover, Monte Carlo experiments are carried out to investigate the capabilities of LMS-like algorithms and the ILL algorithm for FECG extraction. The results demonstrate that the ILL algorithm outperforms the LMS-like ones with carefully selected parameters, particularly showcasing superior robustness against impulsive noise. This work holds significance both in the theoretical research of adaptive filtering and in its practical application for FECG extraction.

This work presents an effective and robust methodology to achieve the joint estimation of direction-of-arrival (DOA) and range. One or more near-field sources are the signals of interest. Far-field sources and alpha-stable noise are additive interferences and will be removed in two steps. First, alpha-stable noise is suppressed into Gaussian or sub-Gaussian noise through a proposed mathematical limiter. Then, the near-field sources are separated from the mixed sources according to the specific technique. The two steps involve three key technologies, including noise reduction, sparse reconstruction, and subspace decomposition. In the experiments, we design an assessment standard for missing detection in data calculation, which is of utmost importance to demonstrate the novel methodology's superiority compared to the other existing competitive ones

Abstract Hierarchical text classification (HTC) is an important task with broad applications, and few-shot HTC has gained increasing interest recently. While in-context learning (ICL) with large language models (LLMs) has achieved significant success in few-shot learning, it is not as effective for HTC because of the expansive hierarchical label sets and extremely ambiguous labels. In this work, we introduce the first ICL-based framework with LLM for few-shot HTC. We exploit a retrieval database to identify relevant demonstrations, and an iterative policy to manage multi-layer hierarchical labels. Particularly, we equip the retrieval database with HTC label-aware representations for the input texts, which is achieved by continual training on a pretrained language model with masked language modeling (MLM), layer-wise classification (CLS, specifically for HTC), and a novel divergent contrastive learning (DCL, mainly for adjacent semantically similar labels) objective. Experimental results on three benchmark datasets demonstrate superior performance of our method, and we can achieve state-of-the-art results in few-shot HTC.

Between 2018 and 2022, INRAE Bordeaux (IB) has developed a series of soil moisture (SM) and L-band Vegetation Optical depth (L-VOD) retrieval products from SMOS and SMAP, which are currently the only two operational L-band passive microwave satellite missions. These IB products rely on a two-parameter inversion of the L-MEB model (L-band Microwave Emission of the Biosphere) which requires little ancillary information. The products are found to be accurate, and very well-suited for application in hydrology, agriculture, climate and vegetation monitoring. In this communication, we present an overview of the development, evaluation and new applications of these IB SM or L-VOD products.

The scaling effect in remote sensing limits the estimation accuracy and application of remote sensing products, such as leaf area index (LAI). At present, the studies on scaling effect mostly focus on developing the algorithm for scaling bias correction, and rarely consider the model parameter type, which plays a critical role during the scaling bias calculation of LAI. In addition, some studies have suggested that the nonlinearity of normalized difference vegetation index (NDVI) equation affects the scaling bias calculation of LAI. However, few studies have been performed to clarify its effects in a quantitative way. In this paper, at two VALERI sites, discuss on the scaling bias calculation of LAI based on the Taylor series expansion method (TSEM) from two aspects: i) the nonlinearity of NDVI equation and ii) the model parameter selection (reflectance, NDVI and directional gap probability). The results indicate that the nonlinearity of NDVI equation has little effects on the LAI scaling bias calculation when NDVI was included in the retrieval model. On the other hand, when directional gap probability was considered in the retrieval model, it had effects. In comparison, when directional gap probability was used in the retrieval model, the scaling bias calculation of LAI showed higher quality.

Vegetation productivity is an important parameter for estimating carbon stocks in terrestrial ecosystems and is important for monitoring regional and global ecological changes. In this study, gross primary productivity (GPP) and net primary productivity (NPP) products with a spatial resolution of 500 m and a temporal resolution of 8 days from 2000 to 2019 were produced based on Global land surface satellite (GLASS) leaf area index (LAI) and the fraction of absorbed photosynthetically active radiation (FPAR) products, and an improved light use efficiency (LUE) model that introduced clearness index (CI) to represent the effect of radiation on LUE. Validated by FLUXNET GPP data, Bigfoot NPP and EMID NPP data, the GPP and NPP products have high accuracy. The dataset has the potential to monitor global and regional ecology and vegetation growth conditions.

Interannual variability in biomass represented by SMOS vegetation optical depth (VOD) and precipitation was assessed over the Contiguous United States. The greatest interannual variability in both VOD and precipitation occurred in shrubs and herbaceous (grasslands), with forests the least variable. At a continental scale, VOD was strongly correlated with annual precipitation. Results showed a significant correlation coefficient (∼ 0.93) between interannual variability of precipitation and biomass, indicating that the interannual variability of precipitation could be a good predictor of the interannual variability of biomass.

The present study focuses on retrieving soil and canopy temperatures, which are key parameters to estimate soil moisture and vegetation optical depth from multi-frequency microwaves information. Several retrieval algorithms assume that canopy and vegetation temperatures are similar in thermal equilibrium conditions, while others separate their contributions, as SMOS-IC, one of the consolidated retrieval algorithms for the Soil Moisture and Ocean Salinity (SMOS) satellite mission. Soil and canopy temperatures in SMOS-IC are modelled from the ECMWF (European Centre for Medium-Range Weather Forecasts) centre. Both SMOS and the Soil Moisture Active Passive (SMAP) missions are currently the only passive L-band (1.4 GHz) missions in operation, but their lifetime is limited. In this context, the upcoming Copernicus Imaging Microwave Radiometer (CIMR) mission will provide continuity on L-band measurements with complementary information in a range of microwave frequencies, from 1.4 to 36.5 GHz. This study uses in situ soil moisture information from the International Soil Moisture Network (ISMN) as input in the SMOS-IC algorithm to retrieve vegetation optical depth (VOD) and soil/canopy effective temperature (TGC). The retrieved effective temperature is then compared with modelled temperatures from ECMWF and with data from the Advanced Microwave Scanning Radiometer 2 (AMSR2), which acquires the higher frequency bands (C, X, Ka, and Ku) present in the future CIMR mission. Results confirm the potential of all high-frequency bands to estimate TGC, with C and X-bands being the most correlated. This study is a first approach to evaluate how microwave multi-frequency information can help modelling soil and canopy temperatures in the SMOS-IC retrieval algorithm, from which the upcoming CIMR mission may benefit.

Passive microwave remote sensing observations at L-band provide key and global information on surface soil moisture (SM) and vegetation optical depth (VOD), which are related to the Earth water and carbon cycles. Only two spaceborne L-band sensors are currently operating: SMOS, launched end of 2009 and thus providing now a 11-year global dataset and SMAP, launched beginning of 2015. To ensure SM and L-VOD data continuity in the event of failure of one of the space-borne SMOS or SMAP sensors, we developed a consistent brightness temperature (TB) record by first producing consistent 40° SMOS and SMAP TB estimates based on SMOS-IC and SMAP enhanced data resp., and then fusing them via linear fusion method. We found that SMOS and SMAP TB are strongly correlated (R > 0.90 over most of the globe) but present a small bias at both the horizontal and vertical polarizations. The preliminary evaluation results show that this bias can be adjusted using a linear fit, but further evaluation procedures are still needed. In the near future, we will develop a long-term time series of SM and L-VOD products based on this merged SMOS-SMAP TB record.

Global and long-term vegetation optical depth (VOD) dataset are very useful to monitor the dynamics of the vegetation features, climate and environmental changes. In this study, the radar-based global ASCAT (Advanced SCATterometer) IB (INRAE-BORDEAUX) VOD was retrieved using a model which was recently calibrated over Africa. In order to assess the performance of IB VOD, the Saatchi biomass and three other VOD datasets (ASCAT V16, AMSR2 LPRM V5 and VODCA LPRM V6) derived from C-band observations were used in the comparison. The preliminary results show that IB VOD has a promising ability to predict biomass$(\mathrm{R}=0.74,\ \text{RMSE} =44.82\ \text{Mg}\ \text{ha}^{-1})$, which is better than V16 VOD$(\mathrm{R}=0.64,\ \text{RMSE} =51.27\ \text{Mg} \text{ha}^{-1})$and VODCA VOD$(\mathrm{R}=0.72,\ \text{RMSE} =47.14\ \text{Mg}\ \text{ha}^{-1})$. Some retrieval issues for IB VOD were found in boreal regions (e.g., Eastern America, Russia). In the future, we will focus on improving our algorithm in those regions, and produce a global and long-term dataset.

Assessment of the model-based surface soil temperature (ST) products is very important for hydrometeorological and ecological applications, as well as model refinements. Distinguished from previous regional validations using only in situ observations from sparse networks, this study focused on the evaluation of model-based ST products by considering ground observations from 15 dense networks worldwide from April 2015 to December 2017 covering a wide range of ground conditions. Four model-based ST products were selected for the assessment, including the Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2), the Goddard Earth Observing System Model version 5 Forward Processing (GEOS-5 FP), the ERA-Interim and its successor, the newly developed ERA5. The results indicate the GEOS-5 ST product slightly outperforms other ST products by showing an averaged ubRMSD of 1.84 K. All model-based ST products underestimate in situ ST with a negative bias. All four model-based ST products are demonstrated to well capture the temporal trends of ground observations with very promising$R$values larger than 0.97. The ERA5 shows visible improvements compared to its predecessor ERA-Interim by exhibiting smaller ubRMSD, absolute bias and larger$R$values. These findings are expected to provide useful suggestions for the enhancement and specific usage of the model-based ST products.

Vegetation Optical Depth (VOD) plays an increasingly important role in studying global carbon, water and energy transformation [1], [2]. This study explores the performance of the X-MEB (X-band microwave emission of the biosphere) model at global scale. Similar to the L-MEB model, the X-MEB model, built by INRAE (Institut national de recherche pour l'agriculture, l'alimentation et l'environnement) Bordeaux, aims to retrieve VOD (referred to as IB X-VOD) at X-band. To avoid the ill-posed problem caused by retrieving two parameters of interest (soil moisture (SM) and VOD) from mono-angular and dual-polarized observations (AMSR2), which are strongly correlated, we used the ERA5 SM product as an input to the X-MEB inversion. At a first step, we produced global IB X-VOD in year 2015 using the parameters (soil roughness and effective scattering albedo) calibrated in the African continent and evaluated the retrieved X-VOD with three vegetation parameters including Above-Ground Biomass (AGB), Leaf Area Index (LAI) and Normalized Difference Vegetation Index (NDVI). The evaluation results indicate X-MEB model has a great potential for global VOD retrievals from AMSR2 satellite data.

Vegetation optical depth (VOD) is used to parameterize microwave extinction effects within the vegetation layer. Many studies have showed VOD presents interesting features for applications in ecology, water and carbon cycles, and VOD is only marginally impacted by signal disturbances and artefacts from atmospheric, cloud and sun illumination effects. As soil moisture (and not VOD) has generally been the main factor of interest in retrieval studies from microwave observations, there is room for improvement in the retrieved VOD products. In this context, INRAE Bordeaux recently developed alternate VOD products from the SMOS, AMSR2 and ASCAT sensors, by addressing specifically the ill-posed problem of retrieving both SM and VOD from observations which may be strongly cross-correlated. Promising results were obtained particularly in terms of spatial correlation of these alternate VOD indices with biomass.

The long time series vegetation productivity products are of great significance to the research of increasing CO2and global changes. In this paper, global net primary productivity (NPP) in 1981-2018 was firstly estimated with Global LAnd Surface Satellite (GLASS) data, ERA-Interim meteorological data and the other variables by using the improved Multi-source data Synergized Quantitative (MuSyQ) NPP algorithm. The average global NPP is 61.0 PgC/yr in 1981-2018, which is in great agreement with the other similar products. The global NPP has shown a significant increase trend, with an annual growth rate of 0.10 PgC/yr over the past 38 years. NPP in the northern hemisphere and southern hemisphere account for 62.0% and 38.0% of the global respectively, both show an increasing trend. The overall increasing trends in NPP are also consistent among most of the biomes.

Since the first version of the SMOS-IC retrieval product was released in 2017, its soil moisture (SM) and L-band Vegetation Optical depth (VOD) retrievals have proven to be a very interesting alternative product for the SMOS mission. This product relies on a two-parameter inversion of the L-MEB model (L-band Microwave Emission of the Biosphere) which is independent of auxiliary data, a key feature making it well-suited for application in hydrology, agriculture, climate, and carbon cycle. This paper describes the development and validation of the most recent SMOS-IC version (V2) soil moisture product. Compared with the previous version (V105), a new constraint was applied on VOD in the cost function which is minimized in the retrieval process. Soil moisture retrievals from SMOS-IC V2 & V105 were inter-compared against the “European Centre for Medium-Range Weather Forecasts” (ECMWF) modelled SM and the “International Soil Moisture Network” (ISMN) in-situ measurements during 2011-2017 over France. It was found that the average retrieval uncertainty of the new version product was lower than that of the old version, particularly when vegetation density increased. The new version of the SMOS-IC soil moisture product will be made available to the public through the CATDS (Centre Aval de Traitements des Données SMOS) website.

Vegetation Optical Depth (VOD) plays an important role in monitoring the earth ecosystems. There are many VOD products released based on different satellites and frequencies. But most of the VOD products are derived from passive microwave data, and very few active VOD products have been released to date. This study investigated retrievals of the active microwave VOD product from C-band ASCAT (Advanced SCATterometer) observations using the water cloud model in large areas. To achieve this, the ASCAT backscatter data and ECMWF soil moisture data were used as inputs to retrieve ASCAT VOD over the whole Africa. The correlation between the retrieved VOD product and proxies of vegetation density (Saatchi biomass) were used to evaluate the model performance.

Decade years of efforts on the retrieval of soil moisture based on radiative transfer model have largely improved the accuracy of soil moisture (SM). This paper focus on the other parameter, namely vegetation optical depth (VOD). We retrieved X-band VOD from AMSR-E and AMSR2 observations by inverting the L-MEB model (Wigneron et al. 2007 [1]) at X-band, considering that SM was known. As SM input to the L-MEB inversion we used the ECMWF SM product. This step avoids correlation between VOD and SM retrievals from the mono-angular AMSR-E observations. In a first step we evaluated the retrieved VOD with the Copernicus Global Land Service (CGLS) LAI. The evaluation results indicate our model has a great potential for VOD retrievals from AMSR-E/2 satellite data.

NPP is calculated to characterize vegetation activity as well as improve our understanding of terrestrial ecosystem carbon cycle. In this study, we estimated a time series of NPP and the spatial and temporal variations from 1982 to 2012 in China. Subsequently, the correlations between the NPP and climate factors (temperature and precipitation) were evaluated to show the responses of vegetation NPP to climate changes. The results showed that NPP in China decreased from southeast to northwest due to the spatial variability of vegetation types and climate characteristics. Annual NPP had a fluctuating increase tendency during our study period with values ranging from 1.92 to 2.73 PgC•a-1, with an annual increase of 0.02 PgC•a-2. In addition, NPP in north China correlates positively with precipitation and negatively correlates with temperature, this is owing to the fact that this region is relatively dry and increasing precipitation extends the growing season of vegetation. In south China the results were the opposite.