Self-assembling peptides are biomedical materials with unique structures that are formed in response to various environmental conditions. Governed by their physicochemical characteristics, the peptides can form a variety of structures with greater reactivity than conventional non-biological materials. The structural divergence of self-assembling peptides allows for various functional possibilities; when assembled, they can be used as scaffolds for cell and tissue regeneration, and vehicles for drug delivery, conferring controlled release, stability, and targeting, and avoiding side effects of drugs. These peptides can also be used as drugs themselves. In this review, we describe the basic structure and characteristics of self-assembling peptides and the various factors that affect the formation of peptide-based structures. We also summarize the applications of self-assembling peptides in the treatment of various diseases, including cancer. Furthermore, the in-cell self-assembly of peptides, termed reverse self-assembly, is discussed as a novel paradigm for self-assembling peptide-based nanovehicles and nanomedicines.

We predict that managers of firms in countries where languages do not require speakers to grammatically mark future events perceive future consequences of earnings management to be more imminent, and therefore, they are less likely to engage in earnings management. Using data from 38 countries where languages differ on how they encode time, we find that accrual-based earnings management and real earnings management are less prevalent where there is weaker time disassociation in the language. Our analysis based on the birthplace information of U.S. firms' CEOs confirms the relation between languages and earnings management. Our study is the first to examine the relation between the grammatical structure of languages and financial reporting characteristics, and it extends the literature on the effect of informal institutions on corporate actions.

Recently, text-to-speech (TTS) models such as FastSpeech and ParaNet have been proposed to generate mel-spectrograms from text in parallel. Despite the advantages, the parallel TTS models cannot be trained without guidance from autoregressive TTS models as their external aligners. In this work, we propose Glow-TTS, a flow-based generative model for parallel TTS that does not require any external aligner. We introduce Monotonic Alignment Search (MAS), an internal alignment search algorithm for training Glow-TTS. By leveraging the properties of flows, MAS searches for the most probable monotonic alignment between text and the latent representation of speech. Glow-TTS obtains an order-of-magnitude speedup over the autoregressive TTS model, Tacotron 2, at synthesis with comparable speech quality, requiring only 1.5 seconds to synthesize one minute of speech in end-to-end. We further show that our model can be easily extended to a multi-speaker setting. Our demo page and code are available at public. 12

We predict that managers of firms in countries where languages do not require speakers to grammatically mark future events perceive future consequences of earnings management to be more imminent, and therefore, they are less likely to engage in earnings management. Using data from 38 countries where languages differ on how they encode time, we find that accrual-based earnings management and real earnings management are less prevalent where there is weaker time disassociation in the language. Our analysis based on the birthplace information of U.S. firms' CEOs confirms the relation between languages and earnings management. Our study is the first to examine the relation between the grammatical structure of languages and financial reporting characteristics, and it extends the literature on the effect of informal institutions on corporate actions.

No abstract