arXiv subjects

Atharva Kulkarni

Publications and source records attributed to Atharva Kulkarni.

36 records · Page 2Linked to original sources

arXiv2023

Adapting the adapters for code-switching in multilingual ASR

Recently, large pre-trained multilingual speech models have shown potential in scaling Automatic Speech Recognition (ASR) to many low-resource languages. Some of these models employ language adapters in their formulation, which helps to improve monolingual performance and avoids some of the drawbacks of multi-lingual modeling on resource-rich languages. However, this formulation restricts the usability of these models on code-switched speech, where two languages are mixed together in the same utterance. In this work, we propose ways to effectively fine-tune such models on code-switched speech, by assimilating information from both language adapters at each language adaptation point in the network. We also model code-switching as a sequence of latent binary sequences that can be used to guide the flow of information from each language adapter at the frame level. The proposed approaches are evaluated on three code-switched datasets encompassing Arabic, Mandarin, and Hindi languages paired with English, showing consistent improvements in code-switching performance with at least 10\% absolute reduction in CER across all test sets.

cs.CL↗

arXiv2023

Exploring Different Time-series-Transformer (TST) Architectures: A Case Study in Battery Life Prediction for Electric Vehicles (EVs)

In recent years, battery technology for electric vehicles (EVs) has been a major focus, with a significant emphasis on developing new battery materials and chemistries. However, accurately predicting key battery parameters, such as state-of-charge (SOC) and temperature, remains a challenge for constructing advanced battery management systems (BMS). Existing battery models do not comprehensively cover all parameters affecting battery performance, including non-battery-related factors like ambient temperature, cabin temperature, elevation, and regenerative braking during EV operation. Due to the difficulty of incorporating these auxiliary parameters into traditional models, a data-driven approach is suggested. Time-series-transformers (TSTs), leveraging multiheaded attention and parallelization-friendly architecture, are explored alongside LSTM models. Novel TST architectures, including encoder TST + decoder LSTM and a hybrid TST-LSTM, are also developed and compared against existing models. A dataset comprising 72 driving trips in a BMW i3 (60 Ah) is used to address battery life prediction in EVs, aiming to create accurate TST models that incorporate environmental, battery, vehicle driving, and heating circuit data to predict SOC and battery temperature for future time steps.

cs.LG↗

arXiv2023

Search for an isotropic gravitational-wave background with the Parkes Pulsar Timing Array

Pulsar timing arrays aim to detect nanohertz-frequency gravitational waves (GWs). A background of GWs modulates pulsar arrival times and manifests as a stochastic process, common to all pulsars, with a signature spatial correlation. Here we describe a search for an isotropic stochastic gravitational-wave background (GWB) using observations of 30 millisecond pulsars from the third data release of the Parkes Pulsar Timing Array (PPTA), which spans 18 years. Using current Bayesian inference techniques we recover and characterize a common-spectrum noise process. Represented as a strain spectrum $h_c = A(f/1 {\rm yr}^{-1})^α$, we measure $A=3.1^{+1.3}_{-0.9} \times 10^{-15}$ and $α=-0.45 \pm 0.20$ respectively (median and 68% credible interval). For a spectral index of $α=-2/3$, corresponding to an isotropic background of GWs radiated by inspiraling supermassive black hole binaries, we recover an amplitude of $A=2.04^{+0.25}_{-0.22} \times 10^{-15}$. However, we demonstrate that the apparent signal strength is time-dependent, as the first half of our data set can be used to place an upper limit on $A$ that is in tension with the inferred common-spectrum amplitude using the complete data set. We search for spatial correlations in the observations by hierarchically analyzing individual pulsar pairs, which also allows for significance validation through randomizing pulsar positions on the sky. For a process with $α=-2/3$, we measure spatial correlations consistent with a GWB, with an estimated false-alarm probability of $p \lesssim 0.02$ (approx. $2σ$). The long timing baselines of the PPTA and the access to southern pulsars will continue to play an important role in the International Pulsar Timing Array.

astro-ph.HE↗

arXiv2023

The gravitational-wave background null hypothesis: Characterizing noise in millisecond pulsar arrival times with the Parkes Pulsar Timing Array

The noise in millisecond pulsar (MSP) timing data can include contributions from observing instruments, the interstellar medium, the solar wind, solar system ephemeris errors, and the pulsars themselves. The noise environment must be accurately characterized in order to form the null hypothesis from which signal models can be compared, including the signature induced by nanohertz-frequency gravitational waves (GWs). Here we describe the noise models developed for each of the MSPs in the Parkes Pulsar Timing Array (PPTA) third data release, which have been used as the basis of a search for the isotropic stochastic GW background. We model pulsar spin noise, dispersion measure variations, scattering variations, events in the pulsar magnetospheres, solar wind variability, and instrumental effects. We also search for new timing model parameters and detected Shapiro delays in PSR~J0614$-$3329 and PSR~J1902$-$5105. The noise and timing models are validated by testing the normalized and whitened timing residuals for Gaussianity and residual correlations with time. We demonstrate that the choice of noise models significantly affects the inferred properties of a common-spectrum process. Using our detailed models, the recovered common-spectrum noise in the PPTA is consistent with a power law with a spectral index of $γ=13/3$, the value predicted for a stochastic GW background from a population of supermassive black hole binaries driven solely by GW emission.

astro-ph.HE↗

arXiv2023

Revisiting Hate Speech Benchmarks: From Data Curation to System Deployment

Social media is awash with hateful content, much of which is often veiled with linguistic and topical diversity. The benchmark datasets used for hate speech detection do not account for such divagation as they are predominantly compiled using hate lexicons. However, capturing hate signals becomes challenging in neutrally-seeded malicious content. Thus, designing models and datasets that mimic the real-world variability of hate warrants further investigation. To this end, we present GOTHate, a large-scale code-mixed crowdsourced dataset of around 51k posts for hate speech detection from Twitter. GOTHate is neutrally seeded, encompassing different languages and topics. We conduct detailed comparisons of GOTHate with the existing hate speech datasets, highlighting its novelty. We benchmark it with 10 recent baselines. Our extensive empirical and benchmarking experiments suggest that GOTHate is hard to classify in a text-only setup. Thus, we investigate how adding endogenous signals enhances the hate speech detection task. We augment GOTHate with the user's timeline information and ego network, bringing the overall data source closer to the real-world setup for understanding hateful content. Our proposed solution HEN-mBERT is a modular, multilingual, mixture-of-experts model that enriches the linguistic subspace with latent endogenous signals from history, topology, and exemplars. HEN-mBERT transcends the best baseline by 2.5% and 5% in overall macro-F1 and hate class F1, respectively. Inspired by our experiments, in partnership with Wipro AI, we are developing a semi-automated pipeline to detect hateful content as a part of their mission to tackle online harm.

cs.CL↗

arXiv2023

Learning and Reasoning Multifaceted and Longitudinal Data for Poverty Estimates and Livelihood Capabilities of Lagged Regions in Rural India

Poverty is a multifaceted phenomenon linked to the lack of capabilities of households to earn a sustainable livelihood, increasingly being assessed using multidimensional indicators. Its spatial pattern depends on social, economic, political, and regional variables. Artificial intelligence has shown immense scope in analyzing the complexities and nuances of poverty. The proposed project aims to examine the poverty situation of rural India for the period of 1990-2022 based on the quality of life and livelihood indicators. The districts will be classified into `advanced', `catching up', `falling behind', and `lagged' regions. The project proposes to integrate multiple data sources, including conventional national-level large sample household surveys, census surveys, and proxy variables like daytime, and nighttime data from satellite images, and communication networks, to name a few, to provide a comprehensive view of poverty at the district level. The project also intends to examine causation and longitudinal analysis to examine the reasons for poverty. Poverty and inequality could be widening in developing countries due to demographic and growth-agglomerating policies. Therefore, targeting the lagging regions and the vulnerable population is essential to eradicate poverty and improve the quality of life to achieve the goal of `zero poverty'. Thus, the study also focuses on the districts with a higher share of the marginal section of the population compared to the national average to trace the performance of development indicators and their association with poverty in these regions.

cs.CL↗

arXiv2023

Characterizing the Entities in Harmful Memes: Who is the Hero, the Villain, the Victim?

Memes can sway people's opinions over social media as they combine visual and textual information in an easy-to-consume manner. Since memes instantly turn viral, it becomes crucial to infer their intent and potentially associated harmfulness to take timely measures as needed. A common problem associated with meme comprehension lies in detecting the entities referenced and characterizing the role of each of these entities. Here, we aim to understand whether the meme glorifies, vilifies, or victimizes each entity it refers to. To this end, we address the task of role identification of entities in harmful memes, i.e., detecting who is the 'hero', the 'villain', and the 'victim' in the meme, if any. We utilize HVVMemes - a memes dataset on US Politics and Covid-19 memes, released recently as part of the CONSTRAINT@ACL-2022 shared-task. It contains memes, entities referenced, and their associated roles: hero, villain, victim, and other. We further design VECTOR (Visual-semantic role dEteCToR), a robust multi-modal framework for the task, which integrates entity-based contextual information in the multi-modal representation and compare it to several standard unimodal (text-only or image-only) or multi-modal (image+text) models. Our experimental results show that our proposed model achieves an improvement of 4% over the best baseline and 1% over the best competing stand-alone submission from the shared-task. Besides divulging an extensive experimental setup with comparative analyses, we finally highlight the challenges encountered in addressing the complex task of semantic role labeling within memes.

cs.CL↗

arXiv2023

ClArTTS: An Open-Source Classical Arabic Text-to-Speech Corpus

At present, Text-to-speech (TTS) systems that are trained with high-quality transcribed speech data using end-to-end neural models can generate speech that is intelligible, natural, and closely resembles human speech. These models are trained with relatively large single-speaker professionally recorded audio, typically extracted from audiobooks. Meanwhile, due to the scarcity of freely available speech corpora of this kind, a larger gap exists in Arabic TTS research and development. Most of the existing freely available Arabic speech corpora are not suitable for TTS training as they contain multi-speaker casual speech with variations in recording conditions and quality, whereas the corpus curated for speech synthesis are generally small in size and not suitable for training state-of-the-art end-to-end models. In a move towards filling this gap in resources, we present a speech corpus for Classical Arabic Text-to-Speech (ClArTTS) to support the development of end-to-end TTS systems for Arabic. The speech is extracted from a LibriVox audiobook, which is then processed, segmented, and manually transcribed and annotated. The final ClArTTS corpus contains about 12 hours of speech from a single male speaker sampled at 40100 kHz. In this paper, we describe the process of corpus creation and provide details of corpus statistics and a comparison with existing resources. Furthermore, we develop two TTS systems based on Grad-TTS and Glow-TTS and illustrate the performance of the resulting systems via subjective and objective evaluations. The corpus will be made publicly available at www.clartts.com for research purposes, along with the baseline TTS systems demo.

cs.CL↗

arXiv2022

Empowering the Fact-checkers! Automatic Identification of Claim Spans on Twitter

The widespread diffusion of medical and political claims in the wake of COVID-19 has led to a voluminous rise in misinformation and fake news. The current vogue is to employ manual fact-checkers to efficiently classify and verify such data to combat this avalanche of claim-ridden misinformation. However, the rate of information dissemination is such that it vastly outpaces the fact-checkers' strength. Therefore, to aid manual fact-checkers in eliminating the superfluous content, it becomes imperative to automatically identify and extract the snippets of claim-worthy (mis)information present in a post. In this work, we introduce the novel task of Claim Span Identification (CSI). We propose CURT, a large-scale Twitter corpus with token-level claim spans on more than 7.5k tweets. Furthermore, along with the standard token classification baselines, we benchmark our dataset with DABERTa, an adapter-based variation of RoBERTa. The experimental results attest that DABERTa outperforms the baseline systems across several evaluation metrics, improving by about 1.5 points. We also report detailed error analysis to validate the model's performance along with the ablation studies. Lastly, we release our comprehensive span annotation guidelines for public use.

cs.CL↗

arXiv2022

When did you become so smart, oh wise one?! Sarcasm Explanation in Multi-modal Multi-party Dialogues

Indirect speech such as sarcasm achieves a constellation of discourse goals in human communication. While the indirectness of figurative language warrants speakers to achieve certain pragmatic goals, it is challenging for AI agents to comprehend such idiosyncrasies of human communication. Though sarcasm identification has been a well-explored topic in dialogue analysis, for conversational systems to truly grasp a conversation's innate meaning and generate appropriate responses, simply detecting sarcasm is not enough; it is vital to explain its underlying sarcastic connotation to capture its true essence. In this work, we study the discourse structure of sarcastic conversations and propose a novel task - Sarcasm Explanation in Dialogue (SED). Set in a multimodal and code-mixed setting, the task aims to generate natural language explanations of satirical conversations. To this end, we curate WITS, a new dataset to support our task. We propose MAF (Modality Aware Fusion), a multimodal context-aware attention and global information fusion module to capture multimodality and use it to benchmark WITS. The proposed attention module surpasses the traditional multimodal fusion baselines and reports the best performance on almost all metrics. Lastly, we carry out detailed analyses both quantitatively and qualitatively.

cs.CL↗

arXiv2021

Novel active micromixer platform based on Lorentz force for lab-on-a-chip application

Mixing in an active micromixer was achieved using Lorentz force-assisted actuation of an enameled copper wire. A single-step template-assisted soft lithography technique was used to construct the mixing chamber. The chamber had a volume of 1.86μl. Application of a Square wave alternating current in tandem with tension in the wire provided the necessary conditions for the resonant oscillation frequency of the wire. The repeatability of the ratio of higher harmonics to the fundamental frequencies of the oscillating wire conforms to standardization of the device fabrication, assembly, and functionality. Simulations and experiments were performed to validate uniform temperature distribution in the mixing chamber. Real-time optical detection of the sample assisted in sensing the completion of chemical reactions in the chamber. Mixing of various aqueous based chemical reactions was performed. It was found that mixing efficiency was greater than 95 percent. Multiple devices were fabricated to show the usability and reproducibility of the system.

physics.app-ph↗

arXiv2021

Study of a Mismatched Parallel Wire Transmission Line and its Potential Applications

In this work a mismatched section of a transmission line is examined for its utility in a bandpass filtering application. A model to analyze the situation is given. It was found that the transfer characteristics of such mismatched transmission line resemble with that of a narrow bandpass filter with steeper roll-off. It has been shown that the center frequency of the response depends on the length of the mismatched section as well as the dielectric material of the line. Further it has been shown that the center frequency can be tuned to a desired value by adjusting the dielectric profile along the length of the line. This was tested on a set of parallel wire transmission lines of three lengths and liquid dielectrics. During the study it was noticed that a sensing mechanism to detect variation in small volumes of the liquid dielectrics with high sensitivity can be developed using mismatched transmission lines.

physics.app-ph↗

arXiv2021

Communicating Visualizations without Visuals: Investigation of Visualization Alternative Text for People with Visual Impairments

Alternative text is critical in communicating graphics to people who are blind or have low vision. Especially for graphics that contain rich information, such as visualizations, poorly written or an absence of alternative texts can worsen the information access inequality for people with visual impairments. In this work, we consolidate existing guidelines and survey current practices to inspect to what extent current practices and recommendations are aligned. Then, to gain more insight into what people want in visualization alternative texts, we interviewed 22 people with visual impairments regarding their experience with visualizations and their information needs in alternative texts. The study findings suggest that participants actively try to construct an image of visualizations in their head while listening to alternative texts and wish to carry out visualization tasks (e.g., retrieve specific values) as sighted viewers would. The study also provides ample support for the need to reference the underlying data instead of visual elements to reduce users' cognitive burden. Informed by the study, we provide a set of recommendations to compose an informative alternative text.

cs.HC↗

arXiv2021

VLSI Systems for signal processing and Communications

The growing advances in VLSI technology and design tools have exponentially expanded the application domain of digital signal processing over the past 10 years. This survey emphasises on the architectural and performance parameters of VLSI for DSP applications such as speech processing, wireless communication, analog to digital converters, etc

eess.SP↗

arXiv2021

L3CubeMahaSent: A Marathi Tweet-based Sentiment Analysis Dataset

Sentiment analysis is one of the most fundamental tasks in Natural Language Processing. Popular languages like English, Arabic, Russian, Mandarin, and also Indian languages such as Hindi, Bengali, Tamil have seen a significant amount of work in this area. However, the Marathi language which is the third most popular language in India still lags behind due to the absence of proper datasets. In this paper, we present the first major publicly available Marathi Sentiment Analysis Dataset - L3CubeMahaSent. It is curated using tweets extracted from various Maharashtrian personalities' Twitter accounts. Our dataset consists of ~16,000 distinct tweets classified in three broad classes viz. positive, negative, and neutral. We also present the guidelines using which we annotated the tweets. Finally, we present the statistics of our dataset and baseline classification results using CNN, LSTM, ULMFiT, and BERT-based deep learning models.

cs.CL↗

arXiv2021

PVG at WASSA 2021: A Multi-Input, Multi-Task, Transformer-Based Architecture for Empathy and Distress Prediction

Active research pertaining to the affective phenomenon of empathy and distress is invaluable for improving human-machine interaction. Predicting intensities of such complex emotions from textual data is difficult, as these constructs are deeply rooted in the psychological theory. Consequently, for better prediction, it becomes imperative to take into account ancillary factors such as the psychological test scores, demographic features, underlying latent primitive emotions, along with the text's undertone and its psychological complexity. This paper proffers team PVG's solution to the WASSA 2021 Shared Task on Predicting Empathy and Emotion in Reaction to News Stories. Leveraging the textual data, demographic features, psychological test score, and the intrinsic interdependencies of primitive emotions and empathy, we propose a multi-input, multi-task framework for the task of empathy score prediction. Here, the empathy score prediction is considered the primary task, while emotion and empathy classification are considered secondary auxiliary tasks. For the distress score prediction task, the system is further boosted by the addition of lexical features. Our submission ranked 1$^{st}$ based on the average correlation (0.545) as well as the distress correlation (0.574), and 2$^{nd}$ for the empathy Pearson correlation (0.517).

cs.CL↗

arXiv2021

An Attention Ensemble Approach for Efficient Text Classification of Indian Languages

The recent surge of complex attention-based deep learning architectures has led to extraordinary results in various downstream NLP tasks in the English language. However, such research for resource-constrained and morphologically rich Indian vernacular languages has been relatively limited. This paper proffers team SPPU\_AKAH's solution for the TechDOfication 2020 subtask-1f: which focuses on the coarse-grained technical domain identification of short text documents in Marathi, a Devanagari script-based Indian language. Availing the large dataset at hand, a hybrid CNN-BiLSTM attention ensemble model is proposed that competently combines the intermediate sentence representations generated by the convolutional neural network and the bidirectional long short-term memory, leading to efficient text classification. Experimental results show that the proposed model outperforms various baseline machine learning and deep learning models in the given task, giving the best validation accuracy of 89.57\% and f1-score of 0.8875. Furthermore, the solution resulted in the best system submission for this subtask, giving a test accuracy of 64.26\% and f1-score of 0.6157, transcending the performances of other teams as well as the baseline system given by the organizers of the shared task.

cs.CL↗

arXiv2021

Experimental Evaluation of Deep Learning models for Marathi Text Classification

The Marathi language is one of the prominent languages used in India. It is predominantly spoken by the people of Maharashtra. Over the past decade, the usage of language on online platforms has tremendously increased. However, research on Natural Language Processing (NLP) approaches for Marathi text has not received much attention. Marathi is a morphologically rich language and uses a variant of the Devanagari script in the written form. This works aims to provide a comprehensive overview of available resources and models for Marathi text classification. We evaluate CNN, LSTM, ULMFiT, and BERT based models on two publicly available Marathi text classification datasets and present a comparative analysis. The pre-trained Marathi fast text word embeddings by Facebook and IndicNLP are used in conjunction with word-based models. We show that basic single layer models based on CNN and LSTM coupled with FastText embeddings perform on par with the BERT based models on the available datasets. We hope our paper aids focused research and experiments in the area of Marathi NLP.

cs.CL↗