Developing AI agents capable of interacting with open-world environments to solve diverse tasks is a compelling challenge. However, evaluating such open-ended agents remains difficult, with current benchmarks facing scalability limitations. To address this, we introduce \textit{Minecraft Universe} (MCU), a comprehensive evaluation framework set within the open-world video game Minecraft. MCU incorporates three key components: (1) an expanding collection of 3,452 composable atomic tasks that encompasses 11 major categories and 41 subcategories of challenges; (2) a task composition mechanism capable of generating infinite diverse tasks with varying difficulty; and (3) a general evaluation framework that achieves 91.5\% alignment with human ratings for open-ended task assessment. Empirical results reveal that even state-of-the-art foundation agents struggle with the increasing diversity and complexity of tasks. These findings highlight the necessity of MCU as a robust benchmark to drive progress in AI agent development within open-ended environments. Our evaluation code and scripts are available at https://github.com/CraftJarvis/MCU.

MOTIVATION: Identifying regulatory elements in various chromosomal regions that influence gene expression is a fundamental challenge in epigenomics, with profound implications for understanding gene regulation and disease mechanisms. The advent of paired single-cell RNA sequencing and single-cell ATAC sequencing has created unprecedented opportunities to address this challenge by enabling simultaneous profiling of gene expression and chromatin accessibility at single-cell resolution. However, the inherent signals between them are weak due to the highly sparse and noisy nature of data. RESULTS: This article proposes single-cell meta-Path based Omics Embedding (scPOEM), a novel embedding method that jointly projects chromatin accessibility peaks and expressed genes into a shared low-dimensional space. By integrating the relationships among peak-peak, peak-gene, and gene-gene interactions, scPOEM assigns closer representations in the embedding space to related peak-gene pairs. Our experiments demonstrate that scPOEM generates stable representations of peaks and genes, outperforms existing methods in recovering biologically meaningful peak-gene regulatory relationships and enables new insights in subgroup and differential analysis of gene regulation. These results highlight its potential to uncover gene regulatory mechanisms and enhance the understanding of transcriptional regulation at single-cell resolution. AVAILABILITY AND IMPLEMENTATION: The source code of scPOEM is available at https://github.com/Houyt23/scPOEM. The datasets can be obtained from the 10× Genomics (https://www.10xgenomics.com/datasets/pbmc-from-a-healthy-donor-granulocytes-removed-through-cell-sorting-10-k-1-standard-1-0-0) and GEO database under access codes GSE194122 and GSE239916.

Pedestrian historical trajectory completion significantly bolsters the predictive accuracy of models. However, traditional statistical models such as Hidden Markov Models (HMM), which focus solely on individual pedestrian trajectories, often fall short in terms of generalization. Conversely, data-driven deep learning approaches demand extensive and meticulous data annotation as well as large datasets. Additionally, leveraging sequential historical data and uncovering the correlation between neighboring pedestrians during absences presents a significant challenge. To address these issues, we introduce a novel trajectory completion method that harnesses prior-posterior knowledge and interactive information, termed CMPT. Our approach commences with the design of a Neighbor Pedestrian Selection module (NPS), adept at identifying neighboring pedestrians through a composite scoring system that evaluates feature similarity and proximity. Subsequently, we employ a Top-Graph Attention Network (T-GAT) to extract multiple correlation sets between preceding and succeeding moments within the scenario. These correlations are then fed into the Markov-Inverse Recovery module (MR), which utilizes prior and posterior insights to flesh out the neighbor influence at the unobserved intervals. Culminating in the Trajectory Reconstruction module (TR), we integrate the completed neighbor influence data with the historical trajectory of the missing pedestrian to finalize the missing trajectory reconstruction. Empirical evidence from our experiments indicates that the Final Distance Error (FDE) of the trajectories completed by CMPT is a commendable 0.30. The source code for CMPT is available from https://github.com/ZYueliang/CMPT-Net.

Self-supervised learning (SSL) is a challenging task in sleep stage classification (SSC) that is capable of mining valuable representations from unlabeled data. However, traditional SSL methods typically focus on single-view learning and do not fully exploit the interactions among information across multiple views. In this study, we focused on a multi-domain view of the same EEG signal and developed a self-supervised multi-view representation learning framework via time series and time-frequency contrasting (MV-TTFC). In the MV-TTFC framework, we built-in a cross-domain view contrastive learning prediction task to establish connections between the temporal view and time-frequency (TF) view, thereby enhancing the information exchange between multiple views. In addition, to improve the quality of the TF view inputs, we introduced an enhanced multisynchrosqueezing transform, which can create high energy concentration TF image views to compensate for the inaccurate representations in traditional TF processing techniques. Finally, integrating temporal, TF, and fusion space contrastive learning effectively captured the latent features in EEG signals. We evaluated MV-TTFC based on two real-world SSC datasets (SleepEDF-78 and SHHS) and compared it with baseline methods in downstream tasks. Our method exhibited state-of-the-art performance, achieving accuracies of 78.64% and 81.45% with SleepEDF-78 and SHHS, respectively, and macro F1-scores of 70.39% with SleepEDF-78 and 70.47% with SHHS.

A dual-channel interleaved analog-to-digital converter (ADC) operating at 320 MS/s is prototyped to validate a fast-converging foreground time calibration algorithm that is independent of ADC offset errors. An input polarity switching technique is introduced to eliminate the impact of sub-ADC offset mismatches during foreground time calibration. After foreground calibration, the signal-to-noise and distortion ratio (SNDR) and spurious free dynamic range (SFDR) are improved by 8.6 and 18.4 dB, respectively. In the sub-ADC design, a comparison functionality is enabled in the digital circuits to prevent metastability and expedite data conversion. The single-channel conversion rates reach 160 MS/s. The ADC is implemented via 40-nm digital CMOS technology, achieving a 52.01 dB signal-to-noise plus distortion ratio (SNDR) at near-Nyquist input while sampling at 320 MS/s. The overall power consumption is 3.65 mW, which includes an on-chip reference buffer and a clock circuit.

Neural architecture search (NAS) has emerged as a hot topic recently which makes artificial intelligence techniques easier to apply and reduces the demand for experts knowledge by generating deep neural network architectures automatically. However, most existing methods for neural architecture search (NAS) heavily rely on underlying black-box controllers to generate potential candidates of network architectures, and suffer from serious problems of lacking interpretability and search efficiency. In this paper, we propose Disentangled Neural Architecture Search (DNAS), which addresses the two issues by adopting disentangled NAS controller and efficient dense-sampling strategy. Specifically, DNAS learns disentangled factors of network architecture by explicitly encouraging the latent factors to be independent. This approach could not only achieve semantic interpretability but also allow us to conveniently identify the promising regions of representations corresponding to high-performance architectures. We further propose a dense-sampling strategy that conducts targeted architecture search within the promising regions to accelerate the searching process. Our DNAS owns several attractive features: 1) it can successfully learn semantic representations of architectures, including operation selection, skip connections, and layer order; 2) it can speed up the process for neural architecture search more than 13× by using dense-sampling and disentangled factors; 3) it can achieve higher accuracy under less computational cost—DNAS achieves state-of-the-art performance of 94.16% on NASBench-101, and 22.7% top-1 test error on ImageNet with 1.6 GPU-days.