Abstract

In this paper, we tackle the complex task of analyzing televised debates, with a focus on a prime time news debate show from India. Previous methods, which often relied solely on text, fall short in capturing the multimedia essence of these debates [27]. To address this gap, we introduce a comprehensive automated toolkit that employs advanced computer vision and speech-to-text techniques for large-scale multimedia analysis. Utilizing state-of-the-art computer vision algorithms and speech-to-text methods, we transcribe, diarize, and analyze thousands of YouTube videos of prime-time television debates in India. These debates are a central part of Indian media but have been criticized for compromised journalistic integrity and excessive dramatization [18]. Our toolkit provides concrete metrics to assess bias and incivility, capturing a comprehensive multimedia perspective that includes text, audio utterances, and video frames. Our findings reveal significant biases in topic selection and panelist representation, along with alarming levels of incivility. This work offers a scalable, automated approach for future research in multimedia analysis, with profound implications for the quality of public discourse and democratic debate. We will make our data analysis pipeline and collected data publicly available to catalyze further research in this domain.

Abstract

In recent studies, the extensive utilization of large language models has underscored the im- portance of robust evaluation methodologies for assessing text generation quality and rel- evance to specific tasks. This has revealed a prevalent issue known as hallucination, an emergent condition in the model where gener- ated text lacks faithfulness to the source and deviates from the evaluation criteria. In this study, we formally define hallucination and pro- pose a framework for its quantitative detection in a zero-shot setting, leveraging our definition and the assumption that model outputs entail task and sample specific inputs. In detecting hallucinations, our solution achieves an accu- racy of 0.78 in a model-aware setting and 0.61 in a model-agnostic setting. Notably, our so- lution maintains computational efficiency, re- quiring far less computational resources than other SOTA approaches, aligning with the trend towards lightweight and compressed models.

Abstract

In today’s dynamic technological landscape, sustainability has emerged as a pivotal concern, especially with respect to architecting Machine Learning enabled Systems (MLS). Many ML models fail in transitioning to production, primarily hindered by uncertainties due to data variations, evolving requirements, and model instabilities. Machine Learning operations (MLOps) offers a promising solution by enhancing adaptability and technical sustainability in MLS. However, MLOps itself faces challenges related to environmental impact, technical maintenance, and economic concerns. Over the years, self-adaptation has emerged as a potential solution to handle uncertainties. This paper introduces a novel approach employing self-adaptive principles integrated into the MLOps architecture through a MAPE-K loop to bolster MLOps sustainability. By autonomously responding to uncertainties, including data, model dynamics, and environmental variations, our approach aims to address the sustainability concerns of a given MLOps pipeline identified by an architect at design time. Further, we implement the method for a Smart City use case to display the capabilities of our approach.

Abstract

Language models often exhibit undesirable behavior, e.g., generating toxic or gender-biased text. In the case of neural language models, an encoding of the undesirable behavior is often present in the model's representations. Thus, one natural (and common) approach to prevent the model from exhibiting undesirable behavior is to steer the model's representations in a manner that reduces the probability of it generating undesirable text. This paper investigates the formal and empirical properties of steering functions, i.e., transformation of the neural language model's representations that alter its behavior. First, we derive two optimal, in the least-squares sense, affine steering functions under different constraints. Our theory provides justification for existing approaches and offers a novel, improved steering approach. Second, we offer a series of experiments that demonstrate the empirical effectiveness of the methods in mitigating bias and reducing toxic generation.

Abstract

Named Entity Recognition (NER) is a use- ful component in Natural Language Process- ing (NLP) applications. It is used in various tasks such as Machine Translation, Summa- rization, Information Retrieval, and Question- Answering systems. The research on NER is centered around English and some other ma- jor languages, whereas limited attention has been given to Indian languages. We analyze the challenges and propose techniques that can be tailored for Multilingual Named Entity Recog- nition for Indian Languages. We present a hu- man annotated named entity corpora of ∼40K sentences for 4 Indian languages from two of the major Indian language families. Addition- ally,we present a multilingual model fine-tuned on our dataset, which achieves an F1 score of ∼0.80 on our dataset on average. We achieve comparable performance on completely unseen benchmark datasets for Indian languages which affirms the usability of our model.

Abstract

The neural mechanisms underlying the comprehension of meaningful sounds are yet to be fully understood. While previous research has shown that the auditory cortex can classify auditory stimuli into distinct semantic categories, the specific contributions of the primary (A1) and the secondary auditory cortex (A2) to this process are not well understood. We used songbirds as a model species, and analyzed their neural responses as they listened to their entire vocal repertoire ((sim )10 types of vocalizations). We first demonstrate that the distances between the call types in the neural representation spaces of A1 and A2 are correlated with their respective distances in the acoustic feature space. Then, we show that while the neural activity in both A1 and A2 is equally informative of the acoustic category of the vocalizations, A2 is significantly more informative of the semantic category of those vocalizations. Additionally, we show that the semantic categories are more separated in A2. These findings suggest that as the incoming signal moves downstream within the auditory cortex, its acoustic information is preserved, whereas its semantic information is enhanced.

Abstract

A system and method for visualizing large graph data using spirals are provided. The method includes (i) generating communities from an input graph data, (ii) configuring nodes in the communities in an ascending order based on a centrality measure, (iii) determining a spiral for the communities by configuring (a) a node with a highest centrality measure at a spiral center and (b) the nodes in a spiral shape based on the ascending order of the centrality measure along the spiral shape, (iv) determining a first super edge for the spiral of the communities, (v) generating a coarse graph using super nodes, the first super edge, and a second super edge, (vi) configuring, using a standard force-directed layout algorithm, corresponding spirals of the communities on the generated coarse graph based on a weight of the second super edge and attractive forces to generate the visualization of the large graph data.

Abstract

The task of automatically detecting syllable stress is a key module in computer-assisted language learning systems. There are numerous studies proposed in the literature for automatic syllable stress detection by using different knowledge-based prosodic features. Also, different statistical machine learning and deep learning models are explored for this task using knowledge-based features. However, the acoustic parameters considered to compute knowledge-based features might not always represent the stress phenomena, hence the knowledge based features are not always suitable for generalization and scalability. Recently, the rapidly emerging self-supervised learning based representations are outperforming the existing state-of-the-art knowledge-based features in all speech applications. Also, these representations allow the models to be built in an end-to-end fashion. In this work, we explore the use of self-supervised representations (Wav2Vec-2.0), for syllable stress detection and compare the performance with state-of-the-art knowledge-based features. Further, we use our recently proposed explicit representation learning framework, modeled by jointly optimizing variational autoencoder (VAE) and DNN for stress detection. We analyze the performance of representation learning framework with two different state-of-the-art classifiers, support vector machines (SVM) and simple deep neural network (DNN). We conduct experiments on two non-native English speakers’ datasets from ISLE corpus i.e., German (GER), and Italian (ITA). From the analysis study, it is observed that the classification accuracy for syllable stress detection using self-supervised representations significantly improved by 3.2% and 2.7% over knowledge-based features in GER and ITA, respectively. From the t-SNE plots, it is observed that the representations learned by explicit representation learning framework with VAE show better discrimination among stressed and unstressed syllables compared with representations learned implicitly with simple DNN.

Abstract

The maritime transportation industry is a prime target for cybersecurity attacks due to the inherent risks of transmitting sensitive information required for its smooth operation. The introduction of digital systems in maritime transport has made them susceptible to a range of cybersecurity threats. These attacks can lead to the unauthorized acquisition and misuse of vulnerable data, such as personal information of crew and passengers, vessel location, route, schedule, and other related information. To address this issue, a new lightweight authentication protocol has been proposed by utilizing drone technology in conjunction with the 5th generation mobile network (5G) communication. The effectiveness of the proposed protocol has been analyzed, which demonstrates its ability to withstand various security attacks while maintaining low communication and computation costs. Furthermore, it successfully meets the security and functionality requirements of anonymity and untraceability. A comprehensive simulation study and simulation using the network simulator (NS3) have been conducted to assess the impact on various network performance parameters for the proposed scheme. In addition, various experiments using machine-learning algorithms for Big data analytics show the efficiency of the proposed scheme in terms of accuracy, precision, recall and F1 score.

Abstract

No Results Found