Abstract

This paper investigates the nuanced distinctions between typical and atypical speech disfluencies, focusing on features captured by self-supervised models. Typical disfluencies are natural, non-pathological irregularities in speech, while atypical disfluencies are linked to speech disorders like stuttering, characterized by more frequent and severe disruptions. Despite progress in automatic disfluency detection, limited research addresses the direct classification of these two types. This study leverages intermediate representations from four pretrained models Wav2Vec2.0, HuBERT, WavLM, and TERA, to analyze and classify typical and stuttered disfluencies. The experiments utilize two novel Indian English datasets, IIITH-IEDE and IIITH-TISA, enabling a comprehensive analysis of disfluency patterns in a linguistically diverse context. Classification experiments with support vector machines (SVM) and convolutional neural networks (CNN) reveal that features from HuBERT’s 5th layer—balancing low-level acoustic and high-level semantic information—achieve a peak F1 score of 0.97. These findings highlight the importance of intermediate layer representations of self-supervised models in distinguishing nuanced speech variations and contribute to robust and interpretable systems for speech disfluency classification.

Abstract

Recent advancements in multimodal large language models (MLLMs) have driven researchers to explore how well these models read data visualizations, e.g., bar charts, scatter plots. More recently, attention has shifted to visual question answering with maps (Map-VQA). However, Map-VQA research has primarily focused on choropleth maps, which cover only a limited range of thematic categories and visual analytical tasks. To address these gaps, we introduce MapIQ, a benchmark dataset comprising 14,706 question-answer pairs across three map types: choropleth maps, cartograms, and proportional symbol maps spanning topics from six distinct themes (e.g., housing, crime). We evaluate multiple MLLMs using six visual analytical tasks, comparing their performance against one another and a human baseline. An additional experiment examining the impact of map design changes (e.g., altered color schemes, modified legend designs, and removal of map elements) provides insights into the robustness and sensitivity of MLLMs, their reliance on internal geographic knowledge, and potential avenues for improving Map-VQA performance.

Abstract

Accurate and complete product descriptions are crucial for e-commerce, yet seller-provided information often falls short. Customer reviews offer valuable details but are laborious to sift through manually. We present PRAISE: Product Review Attribute Insight Structuring Engine, a novel system that uses Large Language Models (LLMs) to automatically extract, compare, and structure insights from customer reviews and seller descriptions. PRAISE provides users with an intuitive interface to identify missing, contradictory, or partially matching details between these two sources, presenting the discrepancies in a clear, structured format alongside supporting evidence from reviews. This allows sellers to easily enhance their product listings for clarity and persuasiveness, and buyers to better assess product reliability. Our demonstration showcases PRAISE's workflow, its effectiveness in generating actionable structured insights from unstructured reviews, and its potential to significantly improve the quality and trustworthiness of e-commerce product catalogs.

Abstract

Molecular property evaluation units are essential components of property-driven drug-discovery pipelines, as their accuracy and generalizability directly influence the success of compound synthesis and optimization. Existing approaches either employ lightweight models optimized for small molecules or heavyweight architectures capable of handling large molecular systems, each with trade-offs in scope and computational cost. Here, we propose and evaluate a hybrid framework that balances these considerations—delivering broad generalization across diverse molecular sizes while maintaining efficiency during both training and inference. Our results demonstrate that this middle-ground strategy achieves performance on par with large-scale predictors for complex targets, yet retains the speed and resource footprint of more compact models, making it well suited for practical drug-discovery workflows.All our results and code are available in our href{https://github.com/rbSparky/MGT-PlusPlus/tree/saigum}{GitHub repository}.

Abstract

Quantum digital signatures ensure unforgeable message authenticity and integrity using quantum principles, offering unconditional security against both classical and quantum attacks. They are crucial for secure communication in high-stakes environments, ensuring trust and long-term protection in the quantum era. Nowadays, the majority of arbitrated quantum signature (AQS) protocols encrypt data qubit by qubit using the quantum one-time pad (QOTP). Despite providing robust data encryption, QOTP is not a good fit for AQS because of its susceptibility to many types of attacks. In this work, we present an efficient AQS protocol to encrypt quantum message ensembles using a distinct encryption technique, the chained controlled unitary operations. In contrast with existing protocols, our approach successfully prevents disavowal and forgery attacks. We hope this contributes to advancing future investigations into the development of AQS protocols .

Abstract

A method for describing, distributing and re-producing specifications of analog IC using AI models, is fulfilled in the ongoing description by (a) creating an AI model to represent the electrical specifications of the analog integrated circuit over a specified range of PVT, input and output loading, and aging conditions,(b) training the AI model with simulation data associated with the analog IC, wherein the simulation inputs include a plurality of instances of PVT (Process, Voltage, Temperature), input/output loading, and aging conditions,(c) electronically enabling peripheral functions with the AI model into an AI model pack, wherein the peripheral functions include a set of computer code modules to extract the AI model, perform user requested operations, calculate specifications of the analog IC on-the-fly and interact with the user, and (d) distributing the encrypted AI model pack to a plurality of users.

Abstract

Virtual Reality (VR) product requires expertise from diverse set of stakeholders. Moving from requirements to design mock-up(s) while building a VR product is an iterative process and requires manual effort. Due to lack of tool support, creating design templates and managing the respective versions turns out to be laborious and difficult. In this paper, we describe VReqDV, a model-driven VR design generation and versioning tool that can address this gap. The tool uses VR meta-model template as a foundation to facilitate a design pipeline. VReqDV can potentially facilitate design generation, design viewing, design versioning, design to require- ments conformity, traceability, and maintenance. It is a step forward in creating a Model-Driven Development pipeline for VR scene design generation. We demonstrate the capabilities of VReqDV using a simple game scene and share our insights for wider adoption by the VR community

Abstract

Drivable Freespace prediction is a fundamental and crucial problem in autonomous driving. Recent works have addressed the problem by representing the entire non-obstacle road regions as the freespace. In contrast our aim is to estimate the driving corridors that are a navigable subset of the entire road region. Unfortunately, existing corridor estimation methods directly assume a BEV centric representation, which is hard to obtain. In contrast, we frame drivable freespace corridor prediction as a pure image perception task, using only monocular camera input. However such a formulation poses several challenges as one doesn’t have the corresponding data for such freespace corridor segments in the image. Consequently, we develop a novel self-supervised approach for freespace sample generation by leveraging future ego trajectories and front-view camera images, making the process of visual corridor estimation dependent on the ego trajectory. We then employ a diffusion process to model the distribution of such segments in the image. However, the existing binary mask based representation for a segment poses many limitations. Therefore, we introduce ContourDiff, a specialized diffusion-based architecture that denoises over contour points rather than relying on binary mask representations, enabling structured and interpretable freespace predictions. We evaluate our approach qualitatively and quantitatively on both nuScenes and CARLA, demonstrating its effectiveness in accurately predicting safe multimodal navigable corridors in the image.

Abstract

Music has long served as a vehicle for political expression, with protest songs playing a central role in articulating dissent and mobilizing collective action. Yet, despite their cultural significance, the linguistic and acoustic signatures that define protest music remain understudied. We present a multimodal computational analysis of protest and non-protest songs spanning multiple decades. Using NLP and audio analysis, we identify the linguistic and musical features that differentiate protest songs. Instead of focusing on classification performance, we treat classification as a diagnostic tool to investigate these features and reveal broader patterns. Protest songs are not just politically charged they are acoustically and linguistically distinct, and we quantify how.

Abstract

SparseLoc is a global localization framework for city-scale scenarios that leverages vision-language foundation models to generate sparse semantic topometric maps in a zero-shot manner. It combines this map representation with a Monte Carlo localization scheme enhanced by a novel late optimization strategy, ensuring improved pose estimation. By constructing compact yet highly discriminative maps and refining localization through a carefully designed optimization schedule, SparseLoc overcomes the limitations of existing topometric localization methods, offering a better solution for global localization at the scale of a city. Our system achieves over a 5× improvement in localization accuracy compared to existing sparse mapping techniques. Despite utilizing only 1/500th of the points of dense mapping methods, it achieves comparable performance, maintaining an average global localization error below 5m and 2° on KITTI Sequences.