Search

Human groups are able to converge to more accurate beliefs through deliberation, even in the presence of polarization and partisan bias --- a phenomenon known as the ``wisdom of partisan crowds.'' Large Language Models (LLMs) are increasingly being used to simulate human collective behavior, yet few benchmarks exist for evaluating their dynamics against the behavior of human groups. In this paper, we examine the extent to which the wisdom of partisan crowds emerges in groups of LLM-based agents that are prompted to role-play as partisan personas (e.g., Democrat or Republican). We find that they not only display human-like partisan biases, but also converge to more accurate beliefs through deliberation, as humans do. We then identify several factors that interfere with convergence, including the use of chain-of-thought prompting and lack of details in personas. Conversely, fine-tuning on human data appears to enhance convergence. These findings show the potential and limitations of LLM-based agents as a model of human collective intelligence.

Accurately simulating human opinion dynamics is crucial for understanding a variety of societal phenomena, including polarization and the spread of misinformation. However, the agent-based models (ABMs) commonly used for such simulations often over-simplify human behavior. We propose a new approach to simulating opinion dynamics based on populations of Large Language Models (LLMs). Our findings reveal a strong inherent bias in LLM agents towards producing accurate information, leading simulated agents to consensus in line with scientific reality. This bias limits their utility for understanding resistance to consensus views on issues like climate change. After inducing confirmation bias through prompt engineering, however, we observed opinion fragmentation in line with existing agent-based modeling and opinion dynamics research. These insights highlight the promise and limitations of LLM agents in this domain and suggest a path forward: refining LLMs with real-world discourse to better simulate the evolution of human beliefs.

Self-vibrating systems obtaining energy from their surroundings to sustain motion can offer great potential in micro-robots, biomedicine, radar systems, and amusement equipment owing to their adaptability, efficiency, and sustainability. However, there is a growing need for simpler, faster-responding, and easier-to-control systems. In the study, we theoretically present an advanced light-actuated liquid crystal elastomer (LCE) fiber–mass system which can initiate self-sliding motion along a rigid circular track under constant light exposure. Based on an LCE dynamic model and the theorem of angular momentum, the equations for dynamic control of the system are deduced to investigate the dynamic behavior of self-sliding. Numerical analyses show that the theoretical LCE fiber–mass system operates in two distinct states: a static state and a self-sliding state. The impact of various dimensionless variables on the self-sliding amplitude and frequency is further investigated, specifically considering variables like light intensity, initial tangential velocity, the angle of the non-illuminated zone, and the inherent properties of the LCE material. For every increment of π/180 in the amplitude, the elastic coefficient increases by 0.25% and the angle of the non-illuminated zone by 1.63%, while the light intensity contributes to a 20.88% increase. Our findings reveal that, under constant light exposure, the mass element exhibits a robust self-sliding response, indicating its potential for use in energy harvesting and other applications that require sustained periodic motion. Additionally, this system can be extended to other non-circular curved tracks, highlighting its adaptability and versatility. [ABSTRACT FROM AUTHOR]

Background: Immune checkpoint blockade has emerged as a key strategy to the therapy landscape of non‐small cell lung cancer (NSCLC). However, notable differences in immunotherapeutic outcomes exist between the two primary NSCLC subtypes: lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC). This disparity may stem from the tumor immune microenvironment's heterogeneity at the transcriptome level. Methods: By integrative analysis of transcriptomic characterization of 38 NSCLC patients by single‐cell RNA sequencing, the present study revealed a distinct tumor microenvironment (TME) between LUAD and LUSC, with relevant results further confirmed in bulk transcriptomic and multiplex immunofluorescence (mIF) validation cohort of neoadjuvant immunotherapy patients. Results: LUAD exhibited a more active immune microenvironment compared to LUSC. This included highly expression of HLA I/II in cancer cells, reinforced antigen presentation potential of dendritic cells and enhanced cytotoxic activity observed in T/NK cells. In LUSC, cancer cells highly expressed genes belonging to the aldo‐keto reductases, glutathione S‐transferases and aldehyde dehydrogenase family, negatively correlating with immunotherapy outcomes in the validation cohort of our center. Further analysis revealed elevated infiltrated cancer‐associated fibroblasts (CAFs) in LUSC, which was corroborated in The Cancer Genome Atlas cohort. Corresponding increased infiltration of ADH1B+ CAFs in major pathologic response (MPR) patients and the higher presence of FAP+ CAFs in non‐MPR patients were demonstrated by multiplex mIF. Moreover, upregulating immunosuppressive extracellular matrix remodeling was identified in LUSC. Conclusions: These comprehensive analyses advance the understanding of the differences in TME between LUAD and LUSC, offering insights for patient selection and developing subtype‐specific treatment strategies. [ABSTRACT FROM AUTHOR]

The detection probability of underwater weak targets using active sonar is low, and inter-pulse coherent integration can improve the signal-to-noise ratio of echoes. When a target executes a maneuvering turn, complex range and Doppler frequency migrations occur during the coherent integration time that decrease the coherent integration gain. Most existing integration methods simplify the target motion to a finite-order polynomial model but fail to integrate a maneuvering turning target (MTT) due to model mismatch. Hence, this study proposes an underwater MTT integration method based on the modified Radon-Fourier transform. The proposed method constructs a theoretically accurate motion model for the MTT and a phase compensation function to compensate for the Doppler frequency migration. Furthermore, it yields a well-focused integration peak in the range-velocity and offset angle-turn rate dimensions and accurately estimates the target motion parameters. Moreover, the proposed method is suitable for targets with radial and oblique uniform motions. The effectiveness of the proposed method is demonstrated through simulations and a lake test. The proposed method demonstrates good integration performance, with an integration gain approximately 4–7 dB higher than that of traditional methods when using 30 integration pulses. [ABSTRACT FROM AUTHOR]

The northeastern periphery of the Qaidam Basin is a crucial region for comprehending the northeastward growth of the Tibetan Plateau, as it documents Late Cenozoic crustal deformation that elucidates the plateau's growth process. In this study, we reconstruct three stages of crustal deformation in the northeastern periphery of the Qaidam Basin during the Late Cenozoic based on interpretation of growth strata from five seismic profiles, structural mapping of the typical superimposed folds and detailed detrital zircon analysis within the study area. (1) During the Early Miocene to Late Miocene period (23–8.6 Ma), there was NW–SE extensional deformation in the northeast margin of the Qaidam Basin, which exerted control over the deposition of the Youshashan Formation. (2) The NW–SE shortening occurred during the Late Miocene period (8.6–8.1 Ma), subsequent to the deposition of the Youshashan Formation and preceding the deposition of the Shizigou Formation, resulting in a parallel unconformity between these two geological units. (3) The intense shortening of the NE–SW direction occurred during the Late Miocene and Pliocene epochs (8.1–2.5 Ma). The timing of this deformation aligns with the sedimentary age of the Shizigou Formation, suggesting that the initial deformation age may represent the onset of NE extrusion from the Tibetan Plateau towards the northeast margin of the Qaidam Basin. The present study not only delineates a Late Cenozoic structural dome resulting from two‐stage crustal shortening in the northeastern periphery of the Tibetan Plateau, but also provides a crucial evidence for reconstructing the Late Cenozoic intracontinental deformation process in this region. [ABSTRACT FROM AUTHOR]

Vine tea (Ampelopsis grossedentata), a traditional Chinese tea, is rich in flavonoids with various biological activities. Our study found that Vine tea total flavonoids (TFs) treatment reduced the body mass and blood lipid levels and improved the hepatic tissue morphology in mice fed the high‐fat diet (HFD). In vivo, TF treatment activated the hepatic adenosine monophosphate‐activated protein kinase (AMPK)/mammalian target of rapamycin (mTOR) pathway, initiated autophagy, and regulated the expression levels of proteins for lipid metabolism in those HFD‐fed mice. In vitro, TF treatment dramatically reduced the lipid droplets and triacylglycerol content in HepG2 and L02 cells treated with oleic acid (OA). These were associated with the activation of the AMPK/mTOR pathway and autophagy initiation in OA‐treated hepatocytes. This phenotype was abolished in the presence of 3‐methyladenine, an autophagy inhibitor. Our results indicated that the TF activation of AMPK/mTOR leads to the stimulation of autophagy and a decrease in the buildup of intracellular lipids in hepatocytes, showing the potential of TF as a therapeutic agent for nonalcoholic fatty liver disease. Practical Application: Vine tea, a tea drink, has been consumed by Chinese folk for over a thousand years. The result of this study will provide evidence that vine tea total flavonoids have potential use as a functional material for the prevention and amelioration of nonalcoholic fatty liver disease. [ABSTRACT FROM AUTHOR]

Precisely segmenting the organs at risk (OARs) in computed tomography (CT) plays an important role in radiotherapy's treatment planning, aiding in the protection of critical tissues during irradiation. Renowned deep convolutional neural networks (DCNNs) and prevailing transformer-based architectures are widely utilized to accomplish the segmentation task, showcasing advantages in capturing local and contextual characteristics. Graph convolutional networks (GCNs) are another specialized model designed for processing the nongrid dataset, e.g., citation relationship. The DCNNs and GCNs are considered as two distinct models applicable to the grid and nongrid datasets, respectively. Motivated by the recently developed dynamic-channel GCN (DCGCN) that attempts to leverage the graph structure to enhance the feature extracted by the DCNNs, this paper proposes a novel architecture termed adaptive sparse GCN (ASGCN) to mitigate the inherent limitations in DCGCN from the aspect of node's representation and adjacency matrix's construction. For the node's representation, the global average pooling used in the DCGCN is replaced by the learning mechanism to accommodate the segmentation task. For the adjacency matrix, an adaptive regularization strategy is leveraged to penalize the coefficient in the adjacency matrix, resulting in a sparse one that can better exploit the relationships between nodes. Rigorous experiments on multiple OARs' segmentation tasks of the head and neck demonstrate that the proposed ASGCN can effectively improve the segmentation accuracy. Comparison between the proposed method and other prevalent architectures further confirms the superiority of the ASGCN. [ABSTRACT FROM AUTHOR]

In the construction of the civil engineering infrastructure, the noise and vibration are affected by the geological conditions, adopting specific construction techniques based on the geological conditions is of great significance in suppressing the noise and vibration caused by the construction. To classify and evaluate the rock mass quality, the rock quality designation (RQD) is adopted widely in the geological and mining engineering. Traditionally, to obtain RQD, lengths of drilling core pieces are measured and RQD is calculated manually, which is labor-expensive and time-consuming. With the development of the computational power, the image treatment driven by the computer vision creates a potential approach to obtain RQD automatically. In the present work, the image treatment process with the aid of the object detection and the image segmentation is adopted to obtain RQD automatically, based on the similarity of features such as color and texture, the segment anything model is adopted to detect the rock cores in the image, then, the YOLOv8 algorithm is adopted to train the model, and the gap features of the rock chip segments are extracted for segmentation of different rock core segments. To test the performance of the proposed approach, 10 boreholes from Shapingba Railway Comprehensive Reconstruction Project are adopted to conduct the case study. Compared to the traditional manual approach, RQD obtained by the proposed approach is relatively accurate and obviously efficient, namely, the average error is less than 5% and the time consumed is less than 70%. [ABSTRACT FROM AUTHOR]

Accurately estimating the gas-flow rate at a wellhead to invert the formation pressure and production capacity information can be the basis for subsequent well-killing parameter design following oil and gas-well drilling blowout and ignition. Based on the multicomponent characteristics of the blowout gas and the turbulence intensity of the blowout flame, as well as the effects of complex factors such as environmental wind direction, wind speed, and wellhead structure, a numerical model for actual drilling blowout ignition is established. Jet-flame experiments are conducted under blowout conditions to verify the accuracy of the model. The temperature and radiation fields and flame morphologies of the well jet flow flame under different lateral wind speeds, well jet flow rates, and wellhead diameters are analyzed. Previous studies have found that as the lateral wind speed increases, the maximum temperature and maximum thermal radiation intensity of the blowout flame first decrease and then increase. However, as the amount of well jet increases, although the flame influence area increases, the maximum temperature does not increase significantly, and the maximum thermal radiation intensity actually decreases. Based on experimental and numerical simulation datasets, a high-yield gas-well jet volume prediction method based on the flame height and thermal radiation intensity of the well jet flow is constructed, which has important engineering application value for achieving successful well killing following well jet ignition. [ABSTRACT FROM AUTHOR]

To improve the accuracy and efficiency of fish behavior assessment, this paper focuses on quantitatively exploring the variations and relationships between different monitoring dimensions. A systematic comparison was conducted between 3D and 2D behavioral factors using an infrared tracing system, during both day and night. Significant differences in swimming distance were observed among the different monitoring methods, as determined by two‐way ANOVA and Tukey's test. A correction was applied to account for the disparities observed in 2D swimming distance, ensuring accurate measurements. These findings present a cost‐effective and efficient approach for obtaining precise 3D distance data. Additionally, a kinematic factor called the "number of U‐turns" was proposed to provide a more intuitive characterization of directional changes in fish swimming. Significant differences were observed between 2D and 3D data, with higher percentages of false U‐turn counts and missing U‐turn counts compared to correct counts in the 2D view. These findings suggest that reducing the monitoring dimension may impact the accurate estimation of swimming motion, potentially resulting in inaccurate outcomes. Finally, the statistical analyses of the non‐linear properties of fractal dimension revealed significant differences among the various monitoring methods. This conclusion has practical implications for biologists and physicists, enabling them to improve the accuracy of behavioral phenotyping for organisms exhibiting 3D motion. [ABSTRACT FROM AUTHOR]

Background: The geographic distribution and host-parasite interaction networks of Sarcocystis spp. in small mammals in eastern Asia remain incompletely known. Methods: Experimental infections, morphological and molecular characterizations were used for discrimination of a new Sarcocystis species isolated from colubrid snakes and small mammals collected in Thailand, Borneo and China. Results: We identified a new species, Sarcocystis muricoelognathis sp. nov., that features a relatively wide geographic distribution and infects both commensal and forest-inhabiting intermediate hosts. Sarcocystis sporocysts collected from rat snakes (Coelognathus radiatus, C. flavolineatus) in Thailand induced development of sarcocysts in experimental SD rats showing a type 10a cyst wall ultrastructure that was identical with those found in Rattus norvegicus from China and the forest rat Maxomys whiteheadi in Borneo. Its cystozoites had equal sizes in all intermediate hosts and locations, while sporocysts and cystozoites were distinct from other Sarcocystis species. Partial 28S rRNA sequences of S. muricoelognathis from M. whiteheadi were largely identical to those from R. norvegicus in China but distinct from newly sequenced Sarcocystiszuoi. The phylogeny of the nuclear 18S rRNA gene placed S. muricoelognathis within the so-called S. zuoi complex, including Sarcocystisattenuati, S. kani, S. scandentiborneensis and S. zuoi, while the latter clustered with the new species. However, the phylogeny of the ITS1-region confirmed the distinction between S. muricoelognathis and S. zuoi. Moreover, all three gene trees suggested that an isolate previously addressed as S. zuoi from Thailand (KU341120) is conspecific with S. muricoelognathis. Partial mitochondrial cox1 sequences of S. muricoelognathis were almost identical with those from other members of the group suggesting a shared, recent ancestry. Additionally, we isolated two partial 28S rRNA Sarcocystis sequences from Low's squirrel Sundasciurus lowii that clustered with those of S. scandentiborneensis from treeshews. Conclusions: Our results provide strong evidence of broad geographic distributions of rodent-associated Sarcocystis and host shifts between commensal and forest small mammal species, even if the known host associations remain likely only snapshots of the true associations. [ABSTRACT FROM AUTHOR]

Polygonum cuspidatum (PC) extract has been listed in the "Catalog of Used Cosmetic Ingredients (2021 Edition)", which can inhibit melanogenesis, thus exerting a whitening effect, and has been widely used in cosmetics. However, there are currently no quality standards for PC extract used in cosmetics, and the bioactive components associated with anti-melanogenesis remain unclear. In view of this, the present study was the first to investigate the spectrum-effect relationship between fingerprints of PC extract and melanogenesis inhibition. Ten batches of PC extract fingerprints were established by HPLC. Pearson's correlation analysis, gray correlation analysis (GRA) and orthogonal partial least squares regression analysis (OPLSR) were used to screen out resveratrol, emodin and physcion as the main whitening active ingredients using the inhibition of tyrosinase in B16F10 cells as the pharmacological index. Then, the melanogenesis inhibitory effects of the above three components were verified by tyrosinase inhibition and a melanin content assay in B16F10 cells. The interaction between small molecules and proteins was investigated by the molecular docking method, and it was confirmed by quantitative real-time PCR (qRT-PCR) that resveratrol, emodin and physcion significantly down-regulated the transcript levels of melanogenesis-related factors. In conclusion, this study established a general model combining HPLC fingerprinting and melanogenesis inhibition and also analyzed the spectrum–effect relationship of PC extract, which provided theoretical support for the quality control of PC extract in whitening cosmetics. [ABSTRACT FROM AUTHOR]

A reduced graphene oxide encapsulating Fe6Ni20Co2Mn2Cu1.5@rGO catalyst is prepared using a Joule heating strategy. The graphene-coated layer with high crystallinity enhances the stability of the crystal structure, resulting in superior OER activity. Rechargeable zinc-air batteries with Fe6Ni20Co2Mn2Cu1.5@rGO demonstrate remarkable performance, boasting a high specific capacity of 800 mA h gZn−1, an impressive peak power density of 154.612 mW cm−2, and a cycle life of 300 hours at a current density of 10 mA cm−2. [ABSTRACT FROM AUTHOR]