Menlo School, also referred to as Menlo, is a private college preparatory school in Atherton, California, United States, across the street from Menlo Park. Menlo comprises a middle school, grades 6–8, with approximately 230 students, and a high school, grades 9–12, with about 570 students. The two schools are located in close physical proximity but operate as semi-autonomous units with select overlapping administration. Menlo was established in 1915, and in 1927 added a junior college that became Menlo College. The college was formally separated from Menlo School in 1994, but they continued to share a single dining hall until 2017. Menlo School is accredited by the Western Association of Schools and Colleges and is a member of the National and California Associations of Independent Schools..
We present a sample of 227 broad-line active galactic nuclei (BLAGNs), incorporating 148 newly identified sources, spanning a redshift interval from z = 0.8 to 7.2. We analyze spectroscopic data from the NIRSpec instrument on board the James Webb Space Telescope, using the G140H, G140M, G235H, G235M, G395H, and G395M gratings to survey N 80,000 galaxies for BLAGNs. Through emission-line fitting, using a sum of Gaussian models for Hα, Hβ, [N II] λλ6548, 6584, and [O III] λλ4959, 5007, we separate active galactic nucleus (AGN) broad-line components from narrow-line emission. We find the detection rate of BLAGNs to be relatively consistent across our redshift range. Compared to typical low-z AGNs (z ≲ 1), high-z BLAGNs are systematically fainter and less massive, yet they accrete more efficiently, with most showing Eddington ratios between 0.1 and 1.0. This confirms rapid black hole growth during the early cosmic epochs. The detection of faint, low-mass BLAGNs at high redshift also helps bridge the observational gap between local supermassive black holes and remote luminous quasars, providing a more complete view of black hole-galaxy coevolution across cosmic time. We also identify 48 known little red dots (LRDs) in our BLAGN sample. For these LRDs, we find a systematic preference for exponential over Gaussian broad-line profiles, which are consistent with electron scattering rather than virial broadening. This implies that standard virial black hole masses for LRDs may be overestimated by up to 2 orders of magnitude.
Part of the longevity strategy of trees is the deposition of extractives, biocidal compounds within heartwood that slow decay long after tree death and are thus an important determinant of carbon residency time. The process of heartwood deposition is regulated by parenchyma rays, living tissues that store carbohydrates, are the sites of biosynthesis, and contain substantial quantities of extractives within mature heartwood. We sought to explore and predict heartwood deposition across the full axial gradient in Sequoia sempervirens by incorporating the plasticity of sapwood parenchyma rays into our models and testing the value of separating them into two size classes. We used synchrotron-based X-ray tomographic microscopy (microCT) for anatomical measurements and novel tissue-level densitometry of wood samples collected from up to 100 m above ground. We found that Sequoia has two distinct ray types, and the interaction of these short and tall rays strongly predicts variation in the extractive content of heartwood. Environmental context was crucial; separate anatomy-based models were needed for never-logged primary and recovering secondary Sequoiaforests in the north and primary forests on the southern margin of the species' range. Incorporating ray structure and distribution into future work on heartwood extractives could provide tools to quantify and monitor heartwood responses of Sequoia forests to management and climate based on the phenotypic plasticity of rays.
This study examines the efficacy of curcumin, a natural polyphenolic compound derived from the turmeric plant, in helping to reduce inflammation and Alzheimer’s disease (AD) symptoms. Three experimental groups of Drosophila were used for testing: GAL4 Drosophila (control group), Drosophila acquiring AD-like symptoms, and Drosophila acquiring AD-like symptoms after curcumin treatment. Two tests were conducted to measure the health of the flies: a climbing assay test and a petri dish performance assay. Both tests use Drosophila’s movement to measure health. This study has developed an AI-based solution with identified behavioral indicators that calculates the speed and height of Drosophila. The Drosophila given curcumin in their diet was relatively more active than the others. The developed AI, OpenCV, calculated results from movement across all three testing groups. The AI OpenCV method used in this project was applied in a novel way by using its image analysis capabilities to measure and compare behavioral or cognitive pattern changes related to curcumin’s potential to delay Alzheimer’s progression, rather than for traditional object detection tasks. The Drosophila given curcumin showcased a significant increase in health, living 30% longer. The Drosophila that acquired Alzheimer-like symptoms with standard fly food died relatively quickly. They showed weaker performance in both assays due to slower mobility and relatively lower health levels. In both tests, the Drosophila that acquired Alzheimer-like symptoms with curcumin in their food moved approximately 5.5 times quicker and had progressively improved health. The results of this study demonstrate that curcumin powder increases speed and movement by 5.5-fold greater than untreated models. Statistical analysis showed highly significant differences between curcumin-treated flies and Drosophila acquiring Alzheimer-like symptoms (p = 0.0086), indicating curcumin’s positive effects. Comparing the control group and the group acquiring Alzheimer-like symptoms also showed high significance (p = 0.0075), confirming successful breeding of Drosophila. There was an insignificant difference between the curcumin group and the control group, suggesting curcumin restored normal health levels. This study illuminate’s curcumin’s efficacy and efficiency in delaying the progression of Alzheimer’s due to its anti-inflammatory and antioxidant properties. This study also demonstrates the immediate effect of an AI-based solution to support the Alzheimer's drug and nutrition study.
We present a measurement of the B-mode polarization power spectrum of the cosmic microwave background anisotropies at 32 <= l < 502 for three bands centered at 95, 150, and 220 GHz using data from the SPT-3G receiver on the South Pole Telescope. This work uses SPT-3G observations from the 2019 and 2020 winter observing seasons of a similar to 1500 deg(2) patch of sky that directly overlaps with fields observed with the BICEP/Keck family of telescopes and covers part of the proposed Simons Observatory and CMB-S4 deep fields. Employing new techniques for mitigating polarized atmospheric noise, the SPT-3G data demonstrates a white noise level of 9.3 (6.7) mu K-arcmin at l similar to 500 for the 95 GHz (150 GHz) data, with a 1/l noise knee at l = 128 (182). We fit the observed six auto- and cross-frequency B-mode power spectra to a model including lensed.CDM B-modes and a combination of Galactic and extragalactic foregrounds. This work characterizes foregrounds in the vicinity of the BICEP/Keck survey area, finding foreground power consistent with that reported by the BICEP/Keck collaboration within the same region and a factor of similar to 3 higher power over the full SPT-3G survey area. Using SPT-3G data over the BICEP/Keck survey area, we place a 95% upper limit on the tensor-to-scalar ratio of r < 0.25 and find the statistical uncertainty on r to be sigma(r) = 0.067.
In this paper, we systemically disentangle BHG, a graph deep learning framework for daily users' ads recommendations. BHG mainly relies on two pillars: (1) graph tokenization to convert the input temporal heterogeneous graph into sequences of tokens, and (2) graph MLP-Mixer neural architecture to learn node representations on sequences of tokens via a mini-batch manner. In general, BHG embraces three advantages: (1) flexibility, i.e., BHG can be seamlessly integrated with any existing industrial recommendation model by treating the learned node embeddings as additional features that encode interactions, (2) efficiency, i.e., the graph tokenization allows sampling the neighborhood both locally and globally, and reduces the number of nodes considered for aggregations, and (3) model simplicity, i.e., the graph MLP-Mixer does not require self-attention for aggregating nodes and hence enjoys the simplicity. We demonstrate the superior performance of the proposed BHG on two internal datasets and one public dataset. We hope this paper can share insights and explain large-scale graph deep learning deployments for researchers, engineers, and practitioners.
