Zero-knowledge succinct non-interactive argument of knowledge (zk-SNARK) serves as a powerful technique for proving the correctness of computations and has attracted significant interest from researchers. Numerous concrete schemes and implementations have been proposed in academia and industry. Unfortunately, the inherent complexity of zk-SNARK has created gaps between researchers, developers and users, as they focus differently on this technique. For example, researchers are dedicated to constructing new efficient proving systems with stronger security and new properties. At the same time, developers and users care more about the implementation's toolchains, usability and compatibility. This gap has hindered the development of zk-SNARK field.
In this work, we provide a comprehensive study of zk-SNARK, from theory to practice, pinpointing gaps and limitations. We first present a master recipe that unifies the main steps in converting a program into a zk-SNARK. We then classify existing zk-SNARKs according to their key techniques. Our classification addresses the main difference in practically valuable properties between existing zk-SNARK schemes. We survey over 40 zk-SNARKs since 2013 and provide a reference table listing their categories and properties. Following the steps in master recipe, we then survey 11 general-purpose popular used libraries. We elaborate on these libraries' usability, compatibility, efficiency and limitations. Since installing and executing these zk-SNARK systems is challenging, we also provide a completely virtual environment in which to run the compiler for each of them. We identify that the proving system is the primary focus in cryptography academia. In contrast, the constraint system presents a bottleneck in industry. To bridge this gap, we offer recommendations and advocate for the open-source community to enhance documentation, standardization and compatibility.
With the rapid development of virtual reality (VR) and augmented reality (AR), spatial audio recording and reproductionhave gained increasing research interest. Higher Order Ambisonics (HOA) stands out for its adaptabilityto various playback devices and its ability to integrate head orientation. However, current HOA recordings oftenrely on bulky spherical microphone arrays (SMA), and portable devices like smartphones are limited by arrayconfiguration and number of microphones. We propose SHB-AE, a spherical harmonic beamforming based methodfor Ambisonics encoding using a smartphone microphone array (SPMA). By designing beamformers for eachorder of spherical harmonic functions based on the array manifold, the method enables Ambisonics encoding andup-scaling. Validation on a real SPMA and its simulated free-field counterpart in noisy and reverberant conditionsshowed that the method successfully encodes and up-scales Ambisonics up to the fourth order with just fourirregularly arranged microphones.
This paper investigates whether a location's growth benefits or suffers from proximity to a big city and explores the underlying mechanisms. Using county-level data from China for 1990–2020, we find that an area's being close to a big city (in the 150–250 km range) reduces its decadal population growth rate by 2.9–3.6 percentage points relative to areas beyond 250 km, which we call the urban growth shadow effect. Initial agricultural employment share has the strongest power to explain whether the negative effect exists. The mechanism is consistent with lower opportunity costs of migration for people employed in agriculture, yet contrasts with core–periphery models that give transport costs a central role. Notably, this effect exhibits a temporal trend. Over time, being proximate to a big city becomes increasingly beneficial.
In response to the growing prevalence of online second language learning and the burgeoning field of international Chinese language education, this study examines the impact of multimodal inputs (MMI) on vocabulary acquisition within online environments among learners of Chinese as a second language (CSL). A teaching intervention was conducted with 90 Mongolian CSL learners, who were grouped into audiovisual, audio, and visual groups. The findings indicate that the audiovisual condition significantly improved vocabulary retention compared to the single-modality conditions in a delayed post-test. Nevertheless, the efficacy of the MMI treatment was observed to vary with learners’ proficiency levels, with beginner-level CSL learners deriving greater benefit from MMI than intermediate-level learners. Furthermore, participants expressed both favorable and critical perspectives regarding the application of MMI in vocabulary instruction. These results highlight the potential of MMI interventions to enhance vocabulary learning in online second-language education, while also underscoring the necessity of considering learners’ target language proficiency and their attitudes when developing MMI-based instructional approaches.
Digital-driven scaling poses significant problems to analog circuits because scaling severely deteriorates transistor current saturation, significantly degrading the intrinsic gain. Special material properties of emerging low-dimensional semiconductors trigger the possibility of providing solutions. We report complementary carbon nanotube thin-film transistors with negative differential resistance-induced current super-saturation for high, exponentially variable intrinsic gain with immunity against degradation during scaling. Current super-saturation at the negative-to-positive differential resistance transition boundary provides intrinsic gain singularities. The large-window, gate-modulated negative differential resistance behavior derived from carbon nanotube’s characteristics enables its practical utilization in circuits. When approaching the singularity, we record that the intrinsic gain varies by orders of magnitude, ranging from 102 to 106 at different operation points. We further demonstrate high and exponentially variable gain in an operational amplifier, showing a tunable single-stage gain ranging from 35 to 60 decibels.
As artificial intelligence-generated content (AIGC) continues to evolve, video-to-audio (V2A) generation has emerged as a key area with promising applications in multimedia editing, augmented reality, and automated content creation. While Transformer and Diffusion models have advanced audio generation, a significant challenge persists in extracting precise semantic information from videos, as current models often lose sequential context by relying solely on frame-based features. To address this, we present TA-V2A, a method that integrates language, audio, and video features to improve semantic representation in latent space. By incorporating large language models for enhanced video comprehension, our approach leverages text guidance to enrich semantic expression. Our diffusion model-based system utilizes automated text modulation to enhance inference quality and efficiency, providing personalized control through text-guided interfaces. This integration enhances semantic expression while ensuring temporal alignment, leading to more accurate and coherent video-to-audio generation.