Searcharxiv⌕ Search

arXiv subjects

Nidhi Sridhar

Publications and source records attributed to Nidhi Sridhar.

2 recordsLinked to original sources

Multimodal Road Closure Mapping via Cross-Modal Graph Attention

Accurate and timely identification of road closures is critical for large-scale navigation and logistics systems, yet remains challenging due to the need to jointly reason over visual evidence, spatial context, and road network topology. Existing approaches either rely on manual map updates or focus on detecting closure-related objects without resolving which road segment is affected. In this work, we introduce a unified multimodal and multitask framework that jointly performs road closure classification (no, partial, complete) and road segment association within a single learning architecture. Our model integrates graph-structured map data, satellite imagery augmented with projected street-level detections, and street-view images using modality-specific encoders and cross-modal transformer attention, enabling explicit reasoning over visual cues and network topology. Evaluated on nearly 20K real-world scenarios collected from a large-scale commercial vehicle fleet, our approach achieves 72.7% closure classification accuracy and 60% road segment association accuracy, outperforming frontier vision-language models by over 28 percentage points on closure classification while being over 4,000x smaller and 100x faster at inference. Extensive ablations demonstrate the importance of cross-modal attention, edge-aware pooling, and graph message passing for accurate geospatial reasoning. This work establishes the first benchmark for joint road closure classification and segment-level attribution, and shows that task-specific architectural inductive biases substantially outperform general-purpose multimodal models for structured geospatial understanding.

cs.CV↗

Shifting Behaviour of Users: Towards Understanding the Fundamental Law of Social Networks

Social Networking Sites (SNSs) are powerful marketing and communication tools. There are hundreds of SNSs that have entered and exited the market over time. The coexistence of multiple SNSs is a rarely observed phenomenon. Most coexisting SNSs either serve different purposes for its users or have cultural differences among them. The introduction of a new SNS with a better set of features can lead to the demise of an existing SNS, as observed in the transition from Orkut to Facebook. The paper proposes a model for analyzing the transition of users from one SNS to another, when a new SNS is introduced in the system. The game theoretic model proposed considers two major factors in determining the success of a new SNS. The first being time that an old SNS gets to stabilise. We study whether the time that a SNS like Facebook received to monopolize its reach had a distinguishable effect. The second factor is the set of features showcased by the new SNS. The results of the model are also experimentally verified with data collected by means of a survey.

cs.SI↗