SONIC: Synergizing VisiON Foundation Models for Stress RecogNItion from ECG signals

Stress recognition through physiological signals such as Electrocardiogram (ECG) signals has garnered significant attention. Traditionally, research in this field predominantly focused on utilizing handcrafted features or raw signals as inputs for learning algorithms. However, there is now a burgeoning interest within the community in leveraging large-scale vision foundation models (VFMs) like ResNet50, VGG19, and others. These VFMs are increasingly preferred due to their ability to capture complex features, enhancing the accuracy and effectiveness of stress recognition systems. However, no particular focus has been given on combining these VFMs. The combination of VFMs offers promising benefits by harnessing their collective knowledge to extract richer representations for improved stress recognition. So, to mitigate this research gap, we focus on combining different VFMs for stress recognition from ECG and propose SONIC, a novel framework that combines VFMs through their logits and training a fully connected network on the combined logits. Through extensive experimentation, SONIC showed the top performance against individual VFMs performance on the WESAD benchmark. With SONIC, we report state-of-the-art (SOTA) performance in WESAD with 99.36% and 99.24% (stress vs non-stress) and 97.66% and 97.10% (amusement vs stress vs baseline) in accuracy and F1 respectively.

Heterogeneity over Homogeneity: Investigating Multilingual Speech Pre-Trained Models for Detecting Audio Deepfake

In this work, we investigate multilingual speech Pre-Trained models (PTMs) for Audio deepfake detection (ADD). We hypothesize that multilingual PTMs trained on large-scale diverse multilingual data gain knowledge about diverse pitches, accents, and tones, during their pre-training phase and making them more robust to variations. As a result, they will be more effective for detecting audio deepfakes. To validate our hypothesis, we extract representations from state-of-the-art (SOTA) PTMs including monolingual, multilingual as well as PTMs trained for speaker and emotion recognition, and evaluated them on ASVSpoof 2019 (ASV), In-the-Wild (ITW), and DECRO benchmark databases. We show that representations from multilingual PTMs, with simple downstream networks, attain the best performance for ADD compared to other PTM representations, which validates our hypothesis. We also explore the possibility of fusion of selected PTM representations for further improvements in ADD, and we propose a framework, MiO (Merge into One) for this purpose. With MiO, we achieve SOTA performance on ASV and ITW and comparable performance on DECRO with current SOTA works.

How Paralingual are Paralinguistic Representations? A Case Study in Speech Emotion Recognition

Pre-trained Models (PTMs) have facilitated substantial progress in the field of Speech Emotion Recognition (SER). SER is an area with applications ranging from HumanComputer Interaction to Healthcare. Recent studies have leveraged various PTM representations as input features for downstream models for SER. PTM specifically pre-trained for paralinguistic tasks have obtained state-of-the-art (SOTA) performance for SER. However, such PTM haven't been evaluated for SER in multilingual settings and experimented only with English. So, we fill this gap, by performing a comprehensive comparative study of five PTMs (TRILLsson, wav2vec2, XLS-R, x-vector, Whisper) for assessing the effectiveness of paralingual PTM (TRILLsson) for SER across multiple languages. Representations from TRILLsson achieved the best performance among all the PTMs. This demonstrates that TRILLsson is able to effectively capture the various paralinguistic features from speech data for better SER. We also show that downstream models using TRILLsson representations achieve SOTA performance in terms of accuracy across various multi-lingual datasets.

A Lightweight Feature Fusion Architecture For Resource-Constrained Crowd Counting

Crowd counting finds direct applications in real-world situations, making computational efficiency and performance crucial. However, most of the previous methods rely on a heavy backbone and a complex downstream architecture that restricts the deployment. To address this challenge and enhance the versatility of crowd-counting models, we introduce two lightweight models. These models maintain the same downstream architecture while incorporating two distinct backbones: MobileNet and MobileViT. We leverage Adjacent Feature Fusion to extract diverse scale features from a Pre-Trained Model (PTM) and subsequently combine these features seamlessly. This approach empowers our models to achieve improved performance while maintaining a compact and efficient design. With the comparison of our proposed models with previously available state-of-the-art (SOTA) methods on ShanghaiTech-A ShanghaiTech-B and UCF-CC-50 dataset, it achieves comparable results while being the most computationally efficient model. Finally, we present a comparative study, an extensive ablation study, along with pruning to show the effectiveness of our models.

From Simulations to Reality: Enhancing Multi-Robot Exploration for Urban Search and Rescue

In this study, we present a novel hybrid algorithm, combining Levy Flight (LF) and Particle Swarm Optimization (PSO) (LF-PSO), tailored for efficient multi-robot exploration in unknown environments with limited communication and no global positioning information. The research addresses the growing interest in employing multiple autonomous robots for exploration tasks, particularly in scenarios such as Urban Search and Rescue (USAR) operations. Multiple robots offer advantages like increased task coverage, robustness, flexibility, and scalability. However, existing approaches often make assumptions such as search area, robot positioning, communication restrictions, and target information that may not hold in real-world situations. The hybrid algorithm leverages LF, known for its effectiveness in large space exploration with sparse targets, and incorporates inter-robot repulsion as a social component through PSO. This combination enhances area exploration efficiency. We redefine the local best and global best positions to suit scenarios without continuous target information. Experimental simulations in a controlled environment demonstrate the algorithm's effectiveness, showcasing improved area coverage compared to traditional methods. In the process of refining our approach and testing it in complex, obstacle-rich environments, the presented work holds promise for enhancing multi-robot exploration in scenarios with limited information and communication capabilities.

Reinforcement Learning-based Knowledge Graph Reasoning for Explainable Fact-checking

Fact-checking is a crucial task as it ensures the prevention of misinformation. However, manual fact-checking cannot keep up with the rate at which false information is generated and disseminated online. Automated fact-checking by machines is significantly quicker than by humans. But for better trust and transparency of these automated systems, explainability in the fact-checking process is necessary. Fact-checking often entails contrasting a factual assertion with a body of knowledge for such explanations. An effective way of representing knowledge is the Knowledge Graph (KG). There have been sufficient works proposed related to fact-checking with the usage of KG but not much focus is given to the application of reinforcement learning (RL) in such cases. To mitigate this gap, we propose an RL-based KG reasoning approach for explainable fact-checking. Extensive experiments on FB15K-277 and NELL-995 datasets reveal that reasoning over a KG is an effective way of producing human-readable explanations in the form of paths and classifications for fact claims. The RL reasoning agent computes a path that either proves or disproves a factual claim, but does not provide a verdict itself. A verdict is reached by a voting mechanism that utilizes paths produced by the agent. These paths can be presented to human readers so that they themselves can decide whether or not the provided evidence is convincing or not. This work will encourage works in this direction for incorporating RL for explainable fact-checking as it increases trustworthiness by providing a human-in-the-loop approach.

Roulette-Wheel Selection-Based PSO Algorithm for Solving the Vehicle Routing Problem with Time Windows

The well-known Vehicle Routing Problem with Time Windows (VRPTW) aims to reduce the cost of moving goods between several destinations while accommodating constraints like set time windows for certain locations and vehicle capacity. Applications of the VRPTW problem in the real world include Supply Chain Management (SCM) and logistic dispatching, both of which are crucial to the economy and are expanding quickly as work habits change. Therefore, to solve the VRPTW problem, metaheuristic algorithms i.e. Particle Swarm Optimization (PSO) have been found to work effectively, however, they can experience premature convergence. To lower the risk of PSO's premature convergence, the authors have solved VRPTW in this paper utilising a novel form of the PSO methodology that uses the Roulette Wheel Method (RWPSO). Computing experiments using the Solomon VRPTW benchmark datasets on the RWPSO demonstrate that RWPSO is competitive with other state-of-the-art algorithms from the literature. Also, comparisons with two cutting-edge algorithms from the literature show how competitive the suggested algorithm is.

Transforming the Embeddings: A Lightweight Technique for Speech Emotion Recognition Tasks

Speech emotion recognition (SER) is a field that has drawn a lot of attention due to its applications in diverse fields. A current trend in methods used for SER is to leverage embeddings from pre-trained models (PTMs) as input features to downstream models. However, the use of embeddings from speaker recognition PTMs hasn't garnered much focus in comparison to other PTM embeddings. To fill this gap and in order to understand the efficacy of speaker recognition PTM embeddings, we perform a comparative analysis of five PTM embeddings. Among all, x-vector embeddings performed the best possibly due to its training for speaker recognition leading to capturing various components of speech such as tone, pitch, etc. Our modeling approach which utilizes x-vector embeddings and mel-frequency cepstral coefficients (MFCC) as input features is the most lightweight approach while achieving comparable accuracy to previous state-of-the-art (SOTA) methods in the CREMA-D benchmark.

A Comparative Study of Pre-trained Speech and Audio Embeddings for Speech Emotion Recognition

Pre-trained models (PTMs) have shown great promise in the speech and audio domain. Embeddings leveraged from these models serve as inputs for learning algorithms with applications in various downstream tasks. One such crucial task is Speech Emotion Recognition (SER) which has a wide range of applications, including dynamic analysis of customer calls, mental health assessment, and personalized language learning. PTM embeddings have helped advance SER, however, a comprehensive comparison of these PTM embeddings that consider multiple facets such as embedding model architecture, data used for pre-training, and the pre-training procedure being followed is missing. A thorough comparison of PTM embeddings will aid in the faster and more efficient development of models and enable their deployment in real-world scenarios. In this work, we exploit this research gap and perform a comparative analysis of embeddings extracted from eight speech and audio PTMs (wav2vec 2.0, data2vec, wavLM, UniSpeech-SAT, wav2clip, YAMNet, x-vector, ECAPA). We perform an extensive empirical analysis with four speech emotion datasets (CREMA-D, TESS, SAVEE, Emo-DB) by training three algorithms (XGBoost, Random Forest, FCN) on the derived embeddings. The results of our study indicate that the best performance is achieved by algorithms trained on embeddings derived from PTMs trained for speaker recognition followed by wav2clip and UniSpeech-SAT. This can relay that the top performance by embeddings from speaker recognition PTMs is most likely due to the model taking up information about numerous speech features such as tone, accent, pitch, and so on during its speaker recognition training. Insights from this work will assist future studies in their selection of embeddings for applications related to SER.

"Can We Detect Substance Use Disorder?": Knowledge and Time Aware Classification on Social Media from Darkweb

Opioid and substance misuse is rampant in the United States today, with the phenomenon known as the "opioid crisis". The relationship between substance use and mental health has been extensively studied, with one possible relationship being: substance misuse causes poor mental health. However, the lack of evidence on the relationship has resulted in opioids being largely inaccessible through legal means. This study analyzes the substance use posts on social media with opioids being sold through crypto market listings. We use the Drug Abuse Ontology, state-of-the-art deep learning, and knowledge-aware BERT-based models to generate sentiment and emotion for the social media posts to understand users' perceptions on social media by investigating questions such as: which synthetic opioids people are optimistic, neutral, or negative about? or what kind of drugs induced fear and sorrow? or what kind of drugs people love or are thankful about? or which drugs people think negatively about? or which opioids cause little to no sentimental reaction. We discuss how we crawled crypto market data and its use in extracting posts for fentanyl, fentanyl analogs, and other novel synthetic opioids. We also perform topic analysis associated with the generated sentiments and emotions to understand which topics correlate with people's responses to various drugs. Additionally, we analyze time-aware neural models built on these features while considering historical sentiment and emotional activity of posts related to a drug. The most effective model performs well (statistically significant) with (macroF1=82.12, recall =83.58) to identify substance use disorder.