Search for: All records

Solution-processable semiconductors hold promise in enabling applications requiring cost-effective electronics at scale but suffer from low performance limited by defects. We show that ordered defect compound semiconductor CuIn₅Se₈, which forms regular defect complexes with defect-pair compensation, can simultaneously achieve high performance and solution processability. CuIn₅Se₈transistors exhibit defect-tolerant, band-like transport supplying an output current above 35 microamperes per micrometer, with a large on/off ratio greater than 10⁶, a small subthreshold swing of 189 ± 21 millivolts per decade, and a high field-effect mobility of 58 ± 10 square centimeters per volt per second, with excellent uniformity and stability, superior to devices built on its less defective parent compound CuInSe₂, analogous binary compound In₂Se₃, and other solution-deposited semiconductors. They can be monolithically integrated with carbon nanotube transistors to form high-speed and low-voltage three-dimensional complementary logic circuits and with micro-light-emitting diodes to realize high-resolution displays. 

This design project arose with the purpose to intervene within the current landscape of content moderation. Our team’s primary focus is community moderators, specifically volunteer moderators for online community spaces. Community moderators play a key role in up-keeping the guidelines and culture of online community spaces, as well as managing and protecting community members against harmful content online. Yet, community moderators notably lack the official resources and training that their commercial moderator counterparts have. To address this, we present ModeratorHub, a knowledge sharing platform that focuses on community moderation. In our current design stage, we focused 2 features: (1) moderation case documentation and (2) moderation case sharing. These are our team’s initial building blocks of a larger intervention aimed to support moderators and promote social support and collaboration among end users of online community ecosystems. 

ABSTRACT Tidal disruption events (TDEs) are routinely observed in quiescent galaxies, as stars from the nuclear star cluster are scattered into the loss cone of the central supermassive black hole (SMBH). TDEs are also expected to occur in active galactic nuclei (AGNs), due to scattering or orbital eccentricity pumping of stars embedded in the innermost regions of the AGN accretion disc. Encounters with embedded stellar-mass black holes (BH) can result in AGN μTDEs. AGN TDEs and μTDEs could therefore account for a fraction of observed AGN variability. Here, by performing scattering experiments with the few-body code SpaceHub, we compute the probability of AGN TDEs and μTDEs as a result of 3-body interactions between stars and binary BHs. We find that AGN TDEs are more probable during the early life of the AGNs, when rates are $$\sim (6\times 10^{-5}-5 \times 10^{-2}) (f_\bullet /0.01)\, \rm {AGN}^{-1}$$ yr−1 (where f• is the ratio between the number density of BHs and stars), generally higher than in quiescent galactic nuclei. By contrast, μTDEs should occur throughout the AGN lifetime at a rate of $$\sim (1\times 10^{-4} - 4\times 10^{-2})(f_\bullet /0.01)\, \rm {AGN}^{-1}$$ yr−1. Detection and characterization of AGN TDEs and μAGN TDEs with future surveys using Rubin and Roman will help constrain the populations of stars and compact objects embedded in AGN discs, a key input for the LVK AGN channel. 

The prevalence and strong capability of large language models (LLMs) present significant safety and ethical risks if exploited by malicious users. To prevent the potentially deceptive usage of LLMs, recent work has proposed algorithms to detect LLM-generated text and protect LLMs. In this paper, we investigate the robustness and reliability of these LLM detectors under adversarial attacks. We study two types of attack strategies: 1) replacing certain words in an LLM’s output with their synonyms given the context; 2) automatically searching for an instructional prompt to alter the writing style of the generation. In both strategies, we leverage an auxiliary LLM to generate the word replacements or the instructional prompt. Different from previous works, we consider a challenging setting where the auxiliary LLM can also be protected by a detector. Experiments reveal that our attacks effectively compromise the performance of all detectors in the study with plausible generations, underscoring the urgent need to improve the robustness of LLM-generated text detection systems. Code is available at https://github.com/shizhouxing/LLM-Detector-Robustness