Although there are several food-level benchmarks for food-related learning, the lack of fine-grained ingredient annotation significantly impedes progress in food scene understanding. In this study, we focus on Chinese food understanding which involves fine-grained ingredient detection and cross-modal ingredient retrieval. Specifically, to support studies on Chinese food understanding, we build the first cross-modal ingredientlevel dataset called CMIngre, which contains 8,001 image-text pairs from three different sources, i.e. dishes, recipes, and usergenerated content, covering 429 distinct ingredients and 95,290 bounding boxes. Based on CMIngre, we evaluate the performance of traditional CNN-based detection algorithms and transformerbased pre-trained large models for ingredient detection. We also propose baseline methods for the cross-modal ingredient retrieval task in both the end-to-end and two-stage settings. Extensive experiments on CMIngre demonstrate the effectiveness of our proposed methods on food understanding

Since rumors widely spread on social networks can cause serious negative impacts, a batch of studies have investigated how to detect rumors. Most of them rely on existing datasets and try to improve the detection performance on those known datasets, but their performance drops significantly when detecting newly emerging events. This is not in line with the original intention of the rumor detection task. To tackle this issue, we formulate the rumor detection problem as a domain adaption task and propose the Class-based Adversarial Domain Adaptive framework, CADA, which is a general model framework where any latest rumor detection methods can be plugged-in. The improvement of new emerging rumor event detection is based on adversarial training. Specifically, CADA considers class-based discriminators to achieve fine-grained alignment of declarations to be detected of different classes. Experiments on three public datasets show that CADA can improve the detection performance of existing rumor detection models and achieve better results than state-of-the-art models. In terms of accuracy, the performance is improved by at least 3% compared with the original base model in the PHEME dataset, and as high as 10% in the Twitter datasets.

Embryonic stem cells (ESCs) consist of a population of self-renewing cells displaying extensive phenotypic and functional heterogeneity. Research towards the understanding of the epigenetic mechanisms underlying the heterogeneity among ESCs is still in its initial stage. Key issues, such as how to identify cell-subset specifically methylated loci and how to interpret the biological meanings of methylation variations remain largely unexplored. To fill in the research gap, we implemented a computational pipeline to analyze single-cell methylome and to perform an integrative analysis with single-cell transcriptome data. According to the origins of variation in DNA methylation, we determined the genomic loci associated with allelic-specific methylation or asymmetric DNA methylation, and explored a beta mixture model to infer the genomic loci exhibiting cell-subset specific methylation (CSM). We observed that the putative CSM loci in ESCs are significantly enriched in CpG island (CGI) shelves and regions with histone marks for promoter and enhancer, and the genes hosting putative CSM loci show wide-ranging expression among ESCs. More interestingly, the putative CSM loci may be clustered into co-methylated modules enriching the binding motifs of distinct sets of transcription factors. Taken together, our study provided a novel tool to explore single-cell methylome and transcriptome to reveal the underlying transcriptional regulatory networks associated with epigenetic heterogeneity of ESCs.

SUMMARY: DMEAS is the first user-friendly tool dedicated to analyze the distribution of DNA methylation patterns for the quantification of epigenetic heterogeneity. It supports the analysis of both locus-specific and genome-wide bisulfite sequencing data. DMEAS progressively scans the mapping results of bisulfite sequencing reads to extract DNA methylation patterns for contiguous CpG dinucleotides. It determines the DNA methylation level and calculates methylation entropy for genomic segments to enable the quantitative assessment of DNA methylation variations observed in cell populations. AVAILABILITY AND IMPLEMENTATION: DMEAS program, user guide and all the testing data are freely available from http://sourceforge.net/projects/dmeas/files/