Description Language Research Articles

Создание языкового контента: теория и практика (на материале веб-версии базы данных «Языки мира» ИЯз РАН)

The article deals with the language description transfer from the encyclopaedic edition «Languages of the World» to the «Languages of the World» database web version. The feature base structure and dependencies within different sections are analysed, along with the possible ways of terminology unification. The article may be of interest to specialists in linguistic typology, data formalisation and development of linguistic databases.

Enhancing semantic mapping in text-to-image diffusion via Gather-and-Bind

Text-to-image synthesis is a challenging task that aims to generate realistic and diverse images from natural language descriptions. However, existing text-to-image diffusion models (e.g., Stable Diffusion) sometimes fail to satisfy the semantic descriptions of the users, especially when the prompts contain multiple concepts or modifiers such as colors. By visualizing the cross-attention maps of the Stable Diffusion model during the denoising process, we find that one of the concepts has a very scattered attention map, which cannot form a whole and gradually gets ignored. Moreover, the attention maps of the modifiers are hard to overlap with the corresponding concepts, resulting in incorrect semantic mapping. To address this issue, we propose a Gather-and-Bind method that intervenes in the cross-attention maps during the denoising process to alleviate the catastrophic forgetting and attribute binding problems without any pre-training. Specifically, we first use information entropy to measure the dispersion degree of the cross-attention maps and construct an information entropy loss to gather these scattered attention maps, which eventually captures all the concepts in the generated output. Furthermore, we construct an attribute binding loss that minimizes the distance between the attention maps of the attributes and their corresponding concepts, which enables the model to establish correct semantic mapping and significantly improves the performance of the baseline model. We conduct extensive experiments on public datasets and demonstrate that our method can better capture the semantic information of the input prompts. Code is available at https://github.com/huan085128/Gather-and-Bind.

Inferential language use in the oral narratives of adolescents with and without developmental language disorder

ABSTRACT Inferential language is an essential aspect of narrative storytelling. It offers not only details about events such as cause and effect and character cognition (e.g. intentions, actions, and reactions) but other evaluative and descriptive language. Previous studies have shown that children with developmental language disorder (DLD, previously also termed as Specific Language Impairment, SLI) exhibit weakness in narrative development, and such weakness is suggested to be unlikely to be resolved over time. However, studies that have focused on the inferential language use in narratives of adolescents with DLD are almost non-existent, and it is unclear whether inferential language use by adolescents with DLD would remain problematic or achieve close-to-normal performance. To address this issue, we compared the use of inferential language in the oral narratives 19 adolescents with DLD (Mean age = 14.3; SD = 0.64) and 19 adolescents with typical language development (TLD; Mean age = 14.0; SD = 0.50). Each transcribed narrative was coded for inferences across character actions/attempts, internal states, causality, character dialogue and other inferences. Results indicated that the two groups performed similarly in all measures except for assigning possessive expressions. Our results suggest that adolescents with DLD may approximate their typical language development (TLD) peers in terms of the inferential language use in oral narratives.

Lexical change is the only constant

Abstract In this study, we look at how language change is presented in two commonly used textbooks for Norwegian lower secondary school. Through thematic content analysis, we look at how linguistic change is presented as opposed to purely synchronic descriptions, as well as how linguistic changes are exemplified. We also look at which language periods are emphasized and how they are discussed. Our findings show that the textbooks mostly discuss languages as static systems and underemphasize how and why languages change. The changes that are presented are overwhelmingly lexical, while phonological, morphological, and syntactic changes are rarely mentioned. The explanations for change provided by the textbooks are mostly external to the linguistic system, such as language contact and politics. Lastly, the descriptions of language tend to be graphocentric. We discuss how this might contribute to a static and normative view of language among students.

A Field-Programmable Gate Array-Based Quasi-Cyclic Low-Density Parity-Check Decoder with High Throughput and Excellent Decoding Performance for 5G New-Radio Standards

This work presents a novel fully parallel decoder architecture designed for high-throughput decoding of Quasi-Cyclic Low-Density Parity-Check (QC-LDPC) codes within the context of 5G New-Radio (NR) communication. The design uses the layered Min-Sum (MS) algorithm and focuses on increasing throughput to meet the strict needs of enhanced Mobile BroadBand (eMBB) applications. We incorporated a Sub-Optimal Low-Latency (SOLL) technique to enhance the critical check node processing stage inherent to the MS algorithm. This technique efficiently computes the two minimum values, rendering the architecture well-suited for specific Ultra-Reliable Low-Latency Communication (URLLC) scenarios. We design the decoder to be reconfigurable, enabling efficient operation across all expansion factors. We rigorously validate the decoder’s effectiveness through meticulous bit-error-rate (BER) performance evaluations using Hardware Description Language (HDL) co-simulation. This co-simulation utilizes a well-established suite of tools encompassing MATLAB/Simulink for system modeling and Vivado, a prominent FPGA design suite, for hardware representation. With 380,737 Look-Up Tables (LUTs) and 32,898 registers, the decoder’s implementation on a Virtex-7 XC7VX980T FPGA platform by AMD/Xilinx shows good hardware utilization. The architecture attains a robust operating frequency of 304.5 MHz and a normalized throughput of 49.5 Gbps, marking a 36% enhancement compared to the state-of-the-art. This advancement propels decoding capabilities to meet the demands of high-speed data processing.

Natural Language Generation for Visualizations: State of the Art, Challenges and Future Directions

AbstractNatural language and visualization are two complementary modalities of human communication that play a crucial role in conveying information effectively. While visualizations help people discover trends, patterns and anomalies in data, natural language descriptions help explain these insights. Thus, combining text with visualizations is a prevalent technique for effectively delivering the core message of the data. Given the rise of natural language generation (NLG), there is a growing interest in automatically creating natural language descriptions for visualizations, which can be used as chart captions, answering questions about charts or telling data‐driven stories. In this survey, we systematically review the state of the art on NLG for visualizations and introduce a taxonomy of the problem. The NLG tasks fall within the domain of natural language interfaces (NLIs) for visualization, an area that has garnered significant attention from both the research community and industry. To narrow down the scope of the survey, we primarily concentrate on the research works that focus on text generation for visualizations. To characterize the NLG problem and the design space of proposed solutions, we pose five Wh‐questions, why and how NLG tasks are performed for visualizations, what the task inputs and outputs are, as well as where and when the generated texts are integrated with visualizations. We categorize the solutions used in the surveyed papers based on these ‘five Wh‐questions’. Finally, we discuss the key challenges and potential avenues for future research in this domain.

An intelligent generative method of fashion design combining attribute knowledge and Stable Diffusion Model

Artificial intelligence generation technology has brought new opportunities to the field of fashion design. Attribute knowledge has a significant impact on the overall effect of fashion design. Contemporary generative methods of fashion design frequently yield results lacking semantic information or missing specific attributes. To address the problem, this study aims for an intelligent generative method of fashion design through constructing prompt templates and a specific attribute low-rank adaption (LoRA) to combine fashion attribute knowledge into the generative process of the Stable Diffusion Model. First, a fashion attribute knowledge graph is constructed to establish prompt templates, and natural language descriptions are transformed into professional and complete prompt through GPT-4. Second, the fashion dataset is annotated with templates to filter specific attributes for LoRA training, followed by controlling fashion attributes in generation. Furthermore, analyses of the generation of women’s jacket designs show that the proposed method consistently improves the accuracy and stability of attributes in fashion design generation.

An Advanced Hardware Implementation of the Trial Division Algorithm for Dividers Utilizing RISC-V ISA

In this paper, we try to design a divider based on RISC-V instruction set architecture (ISA). Since the hardware design of the divider in the ALU is very important for the CPU to implement the arithmetic function. In view of the advantages of the trial division method in the division algorithm and the popularity of the RISC-V ISA, this paper attempts to design a trial division divider utilizing the RISC-V ISA. In this design, the principle of trial division, flow chart, state diagram, RISC-V instructions of division, whole divider module and logic will be designed in detail. To test the function and feasibility, the hardware description is carried out in the Verilog hardware description language (HDL). In this description, four division RISC-V instructions are integrated. Modelsim is used to verify and test the functionality of the divider. The test waveform is used as a result to verify the feasibility of the divider.

Design and Implementation of a 7-Person Voting Device based on Quartus

As the complexity of digital system design continues to increase, efficient and precise decision support systems have become increasingly important. This paper presents the design and implementation of a 7-voter decision system based on Altera’s Quartus software. The voter is applied in the core decision-making process to ensure the accuracy and efficiency of decisions. The study begins by outlining the design principles of the 7-voter decision system, using a design philosophy that minimizes timing delays and optimizes hardware resource utilization. A complex majority voting logic circuit is implemented using high-level hardware description languages (VHDL/Verilog), ensuring the advanced nature and applicability of the voting circuit. In terms of the logical implementation of the voting circuit, the paper conducts an in-depth analysis of the optimization of combinational and sequential logic and standardizes the input and output to enhance the signal stability and the system’s resistance to interference. On the Quartus platform, the design process of the voting system is elaborated in detail, with careful consideration given to boundary conditions and hardware compatibility. The implementation of the voter has undergone rigorous simulation and testing. Strict parameter setting for simulations and waveform analysis confirm the correctness of the circuit, and fault simulation tests further confirm its reliability. Performance indicators such as response time and percentage of resource utilization have been quantitatively analyzed and meet the design objectives set beforehand. Overall, this research not only realizes a highperformance 7-voter decision system but also provides an effective case study for applications on the high-performance Quartus design platform. Future research can build on the current work to further optimize algorithms and hardware, explore the potential for the application of the voter in a broader range of fields, and adapt to the new features of the Quartus platform to achieve more powerful decision support functions.

Design and Implementation of a Synchronous AHB to APB Bridge in System-on-Chip Applications

The requirement of effectively combining key units in an integrated system is proliferating. SoC system is developed to provide a chip-level integration, which become the inevitable trend of integrated circuit development and is widely used in smartphones, industrial applications, and microcontrollers. The ARM AMBA protocol works as a universally adopted way of interaction between various parts of the system. In AMBA architecture, the AHB to APB bridge significantly contributes to combining the high-performance AHB bus and low-power APB bus in the SoC system. This project is to implement an AHB to APB bridge using Verilog, enabling stable data transfer between these two buses. The proposed AHB to APB bridge is intended to fit different read or write strategies and ensure the proper working of the peripherals on the APB bus. The bridge has been implemented via Verilog Hardware description language (HDL). A test bench was created with a virtual AHB host and an optimized SRAM as the high-speed APB peripheral. Verdi simulation shows the bridge completely meets the design intention.

Design and Implementation of Digital Integrated Circuit Based CPU

This paper presents a simplified 16-bit CPU design and its implementation using digital integrated circuits developed with Verilog hardware description language, while emphasizing simulation through the EDA tool. This design puts more emphasis on the critical roles played by the data path and the control unit, executing key instructions including arithmetic operations such as addition, subtraction, bitwise logic, and conditional jumps. This structurally modular CPU is both functioning and stable in nature, allowing support for future enhancements and enhancements as well, thereby serving as an effective model towards an understanding of the principles of CPU design. The execution of these basic instructions puts the core of the CPU under test and reaffirms its proper operation in respect of the detailed instruction set, while its extensibility design confirms further improvement possibilities for future complex instructions as well as finite-state machine optimization for the control unit. The present work not only proves the possibility of creating a fully working CPU, using readily available bare minimum resources to implement a prototype, but is also a breakthrough in future research with CPU architecture. The obtained results will shed light on CPU functionality optimization and extension, being the ground for building more powerful and effective processors in the future.

Flexible Natural Language-Based Image Data Downlink Prioritization for Nanosatellites

Nanosatellites increasingly produce more data than can be downlinked within a reasonable time due to their limited bandwidth and power. Therefore, an on-board system is required to prioritize scientifically significant data for downlinking, as described by scientists. This paper determines whether natural language processing can be used to prioritize remote sensing images on CubeSats with more flexibility compared to existing methods. Two approaches implementing the same conceptual prioritization pipeline are compared. The first uses YOLOv8 and Llama2 to extract image features and compare them with text descriptions via cosine similarity. The second approach employs CLIP, fine-tuned on remote sensing data, to achieve the same. Both approaches are evaluated on real nanosatellite hardware, the VERTECS Camera Control Board. The CLIP approach, particularly the ResNet50-based model, shows the best performance in prioritizing and sequencing remote sensing images. This paper demonstrates that on-orbit prioritization using natural language descriptions is viable and allows for more flexibility than existing methods.

DeLoSo: Detecting Logic Synthesis Optimization Faults Based on Configuration Diversity

Logic synthesis tools are the core components of digital circuit design, which convert programs written in hardware description languages into gate-level netlists, and optimize the netlists. However, the netlist optimization is complex, with numerous optimization parameters to be configured. Any minor optimization faults in logic synthesis tools may cause circuit diagrams to significantly deviate from the original design, posing risks in target systems. We propose DeLoSo, the De tector of Lo gic S ynthesis o ptimization faults, the first method specifically designed to identify potential faults in the optimization processes of logic synthesis tools. DeLoSo relies on netlist differences and parameter variations to guide the generation of diverse Logic Synthesis Optimization Configuration (LSOC) combinations to thoroughly test the optimization process. DeLoSo consists of three components: LSOC generator, which generates diverse LSOC combinations through configuration recombination and mutation; LSOC diversity evaluator, which assesses the diversity of optimization configurations; and LSOC validator, which validates the generated LSOC combinations to discover optimization faults. DeLoSo identified 19 faults in two established logic synthesis tools (i.e., Vivado and Yosys); 15 of them have been fixed by vendors. Particularly, the community maintainers of Yosys have considered incorporating DeLoSo into Yosys’ existing test suite.

GestureGPT: Toward Zero-Shot Free-Form Hand Gesture Understanding with Large Language Model Agents

Existing gesture interfaces only work with a fixed set of gestures defined either by interface designers or by users themselves, which introduces learning or demonstration efforts that diminish their naturalness. Humans, on the other hand, understand free-form gestures by synthesizing the gesture, context, experience, and common sense. In this way, the user does not need to learn, demonstrate, or associate gestures. We introduce GestureGPT, a free-form hand gesture understanding framework that mimics human gesture understanding procedures to enable a natural free-form gestural interface. Our framework leverages multiple Large Language Model agents to manage and synthesize gesture and context information, then infers the interaction intent by associating the gesture with an interface function. More specifically, our triple-agent framework includes a Gesture Description Agent that automatically segments and formulates natural language descriptions of hand poses and movements based on hand landmark coordinates. The description is deciphered by a Gesture Inference Agent through self-reasoning and querying about the interaction context (e.g., interaction history, gaze data), which is managed by a Context Management Agent. Following iterative exchanges, the Gesture Inference Agent discerns the user’s intent by grounding it to an interactive function. We validated our framework offline under two real-world scenarios: smart home control and online video streaming. The average zero-shot Top-1/Top-5 grounding accuracies are 44.79%/83.59% for smart home tasks and 37.50%/73.44% for video streaming tasks. We also provide an extensive discussion that includes rationale for model selection, generalizability, and future research directions for a practical system etc.

NoviCode: Generating Programs from Natural Language Utterances by Novices

Abstract Current Text-to-Code models demonstrate impressive capabilities in generating executable code from natural language snippets. However, current studies focus on technical instructions and programmer-oriented language, and it is an open question whether these models can effectively translate natural language descriptions given by non-technical users and express complex goals, to an executable program that contains an intricate flow—composed of API access and control structures as loops, conditions, and sequences. To unlock the challenge of generating a complete program from a plain non-technical description we present NoviCode, a novel NL Programming task, which takes as input an API and a natural language description by a novice non-programmer, and provides an executable program as output. To assess the efficacy of models on this task, we provide a novel benchmark accompanied by test suites wherein the generated program code is assessed not according to their form, but according to their functional execution. Our experiments show that, first, NoviCode is indeed a challenging task in the code synthesis domain, and that generating complex code from non-technical instructions goes beyond the current Text-to-Code paradigm. Second, we show that a novel approach wherein we align the NL utterances with the compositional hierarchical structure of the code, greatly enhances the performance of LLMs on this task, compared with the end-to-end Text-to-Code counterparts.

Effects of Adjectives and Adverbs on Visually Impaired Students in Depicting Contents and Enhancing Retention in Inclusive Classrooms in Katsina State, Nigeria

This study investigated the impact of adjectives and adverbs on the content depiction and retention of visually impaired students within a Nigerian context. By employing a mixed-methods approach, the research examined the effects of these linguistic elements on students’ comprehension, engagement, and overall learning experience. The findings revealed a significant positive correlation between using adjectives and adverbs and students’ academic performance. Students perceived descriptive language as enhancing their understanding, imagination, and recall of information. Teachers predominantly incorporated adjectives and adverbs in their instruction, aligning with inclusive education principles. However, variations in the frequency and priority of descriptive language usage among teachers were observed. The study concludes that incorporating adjectives and adverbs into teaching practices significantly benefits visually impaired students by improving their content depiction and retention. These findings underscore the importance of descriptive language in creating inclusive and engaging learning environments for all students. Further research is recommended to explore the long-term effects of descriptive language on student outcomes and to develop specific pedagogical strategies for maximizing its impact on visually impaired learners.

Too many numbers and worse word choice: Why readers find data-driven news articles produced with automation harder to understand

The production of data-driven journalism is becoming increasingly automated, impacting its composition and comprehensibility. Given the importance of data-driven reporting for democratic participation, this study investigates, firstly, how readers evaluate the composition of data-driven articles produced with and without the help of automation and, secondly, how these evaluations affect readers’ perceptions of the articles’ comprehensibility. In an online survey experiment, 3135 online news consumers evaluated 24 articles produced with or without automation using criteria developed in prior research. Our factor analysis reduced those criteria to five categories that matter in readers’ evaluations of the articles’ composition: numeric features, writing style, sentence and paragraph length, descriptive language, and word choice. Our results show, firstly, that although the perception of news stories produced with automation did not differ significantly from that of news stories produced without automation regarding sentence and paragraph length and writing style, the stories produced with automation were evaluated as significantly less comprehensible; and, secondly, that this can be explained partly by readers’ perceptions of some of the other article composition categories, which were rated significantly worse in automated articles. Our findings suggest that using automation to produce data-driven news articles changes their perceived composition in ways that negatively impact comprehensibility. However, this study also suggests how such articles could be made more comprehensible. Specifically, when ‘post-editing’ automated articles, journalists should aim to further reduce the quantity of numbers, better explain words that readers are unlikely to understand and change inappropriate wording.

Image Captioning Based on Semantic Scenes.

With the development of artificial intelligence and deep learning technologies, image captioning has become an important research direction at the intersection of computer vision and natural language processing. The purpose of image captioning is to generate corresponding natural language descriptions by understanding the content of images. This technology has broad application prospects in fields such as image retrieval, autonomous driving, and visual question answering. Currently, many researchers have proposed region-based image captioning methods. These methods generate captions by extracting features from different regions of an image. However, they often rely on local features of the image and overlook the understanding of the overall scene, leading to captions that lack coherence and accuracy when dealing with complex scenes. Additionally, image captioning methods are unable to extract complete semantic information from visual data, which may lead to captions with biases and deficiencies. Due to these reasons, existing methods struggle to generate comprehensive and accurate captions. To fill this gap, we propose the Semantic Scenes Encoder (SSE) for image captioning. It first extracts a scene graph from the image and integrates it into the encoding of the image information. Then, it extracts a semantic graph from the captions and preserves semantic information through a learnable attention mechanism, which we refer to as the dictionary. During the generation of captions, it combines the encoded information of the image and the learned semantic information to generate complete and accurate captions. To verify the effectiveness of the SSE, we tested the model on the MSCOCO dataset. The experimental results show that the SSE improves the overall quality of the captions. The improvement in scores across multiple evaluation metrics further demonstrates that the SSE possesses significant advantages when processing identical images.

The Pedagogical Tipiṭaka: OER & the Three Baskets of Ancient Language Instruction

Ancient languages present a unique teaching challenge: for spoken languages, common pedagogy recommends engaging students via dialogue; for ancient languages, no speakers survive with whom to practise. This paper highlights how the Linguistics Research Center at the University of Texas at Austin has approached this challenge by creating the Early Indo-European OnLine (EIEOL) collection, an online educational resource whose lesson series present early languages directly through original, unsimplified ancient texts. Currently accessed by over 20,000 users per month, EIEOL spans 18 languages, from Greek and Latin to Old Church Slavonic, Sanskrit, and other important languages of ancient Asia such as Hittite, Classical Armenian, Avestan, and Tocharian. Each series presents extensively annotated excerpts of original texts in the target language, with accompanying modules explaining grammar and context. The text-centred approach affords learners a direct path to understanding that suits a variety of experience levels and minimises the conceptual grammatical apparatus necessary to begin interpreting original texts. This format fosters theoretical flexibility, adaptable to different approaches and grammatical descriptions of ancient languages. It is also useful for languages whose grammatical structures have shifted dramatically over their history, like Tocharian, or remain hotly debated or under-described by experts. Finally, it facilitates applications to typologically diverse languages and language families, with early Mesoamerican, Semitic, and Sino-Tibetan language series already under development. The EIEOL infrastructure therefore provides a robust platform for free, text-centred, self-paced introductions to ancient languages from a variety of language families.

Hardware-accelerated neural network model for early prediction of sudden cardiac arrest based on heart rate variability metrics

AbstractSudden Cardiac Arrest (SCA) constitutes a dire medical condition, marked by the abrupt cessation of effective blood circulation due to the heart's failure to contract properly. This leads to acute circulatory collapse, often culminating in loss of consciousness within an hour and potentially resulting in fatality within minutes if left unattended. Heart rate variability (HRV) serves as a critical biometric, derived from electrocardiogram (ECG) signals with ventricular depolarization waves to further calculate the R-R Intervals (RRIs). These intervals provide the basis for extracting various characteristics of cardiac rhythm, encompassing time-domain, frequency-domain, and nonlinear features. This study presents a neural network-based classification algorithm that leverages HRV metrics to categorize patients into SCA and Normal Sinus Rhythm (NSR) cohorts. The proposed neural network (NN) model showcased impressive results, achieving an accuracy of 96.23%, a sensitivity of 94.35%, and a specificity of 98.12% in detecting SCA, as evaluated through leave-one-subject-out analysis. In order to harness the benefits of hardware acceleration, the algorithm is implemented on a Field-Programmable Gate Array (FPGA). Its computational efficiency is subsequently benchmarked against traditional software-based methodologies. The hardware-level implementation is made possible in Verilog hardware description language (HDL) and was verified successfully with expected performance by register-transfer level (RTL) simulation via Vivado 2020.2.