Pasadena City College

This article reports on the initial calibration and performance of the High-resolution Volatiles and Minerals Moon Mapper (), slated for launch on the National Aeronautics and Space Administration's Lunar Trailblazer mission. is an imaging spectrometer measuring from 600 to 3,600 nm with 10-nm spectral sampling and 50-90 m/pixel ground sampling. The mission goal is to understand the form, abundance, and distribution of water across the lunar surface and the lunar water cycle, accomplished by measuring the distinct absorptions of water ice, adsorbed O, and OH/hydroxl while controlling for thermal effects with . also has the ability to measure mineralogical composition. has been assembled, tested and calibrated in preparation for launch and integrated on the Lunar Trailblazer spacecraft. We review the design, calibration process, results, and implications for Lunar Trailblazer science goals. We find 's radiometric sensitivity is sufficient to confidently measure 1% differences in absorption band strengths under direct solar illumination in single pixel data. In addition, has the radiometric precision to discriminate different species of volatile absorptions at irradiances of 1 W , which will enable mapping and discriminating water ice or other volatiles within most of the Moon's Permanently Shadowed Regions using terrain-scattered illumination.

Retrieval-augmented generation (RAG) is the backbone of knowledge-intensive NLP, yet its progress is hindered by a long-standing asymmetry: Generators are refined while retrievers remain static, and full end-to-end optimization is prohibitively unstable. We present BPO-RAG, a bi-level preference-learning framework that redefines the training paradigm by jointly optimizing retrieval and generation with a single supervision signal, pairwise preferences. Stage~1 (Retrieval Preference Optimization) learns to select superior evidence sets, while Stage~2 (Generation Preference Optimization) aligns answer generation with the same evidence, closing the gap between what to read and what to write. This recipe without label requires no reward model or online RL, integrates seamlessly with standard RAG pipelines, and transforms preferences into a unifying training currency. Across open-domain QA benchmarks, BPO-RAG consistently advances retrieval quality and yields more accurate, faithful answers, surpassing strong RAG baselines with remarkable stability. By coupling retrieval and generation under a unified preference framework, BPO-RAG establishes a practical and principled path toward the next generation of reliable, modular, and trustworthy knowledge-intensive language models.

Artificial biomolecular condensates are a promising tool to achieve spatial self-organization in active materials and synthetic cells. DNA branched motifs known as nanostars are a particularly versatile tool to build artificial condensates with customizable phase diagrams and viscoelastic properties. Here, we characterize how the inclusion of aptamers in DNA nanostars makes it possible to engineer condensates with the capacity of recruiting biomolecules of interest to the dense phase. Furthermore, we demonstrate how biomolecules can be released from condensates by supplying "kleptamer", an oligonucleotide that is complementary to the aptamer sequence. We focus specifically on incorporating a DNA aptamer that recruits streptavidin (SA), and investigate how the positioning of the aptamer-whether at the nanostar junction, or at the middle or tip of one of the nanostar arms-affects the formation of condensates and the recruitment efficiency of SA. We find that the aptamer's location within the nanostar does not significantly influence condensation and recruitment, nor the capacity of kleptamers to release the protein. These results provide new insight into the design of synthetic DNA condensates for the uptake and release of target molecules and demonstrate their robustness with respect to nanostar design, potentially broadening the use of such condensates in synthetic biology applications.