RoboKA: KAN Informed Multimodal Learning for RoboCall Surveillance System
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914523748761600 |
|---|---|
| author | Choudhury, Nitin Kumar, Nikhil Sinha, Aditya Kumar Anand, Abhijeet Salemi, Hossein Phukan, Orchid Chetia Purohit, Hemant Buduru, Arun Balaji |
| author_facet | Choudhury, Nitin Kumar, Nikhil Sinha, Aditya Kumar Anand, Abhijeet Salemi, Hossein Phukan, Orchid Chetia Purohit, Hemant Buduru, Arun Balaji |
| contents | Wide exploration on robocall surveillance research is hindered due to limited access to public datasets, due to privacy concerns. In this work, we first curate Robo-SAr, a synthetic robocall dataset designed for robocall surveillance research. Robo-SAr comprises of ~200 unwanted and ~1200 legitimate synthetic robocall samples across three realistic adversarial axes: psycholinguistics-manipulated transcripts, emotion-eliciting speech, and cloned voices. We further propose RoboKA, a Kolmogorov-Arnold Network (KAN)-based multimodal fusion framework designed to model structured nonlinear interactions between acoustic and linguistic cues that characterize diverse adversarial robocall strategies. RoboKA first leverages cross-modal contrastive learning to align latent modality representations and feeds the resulting embeddings to a KAN-projection head for final classification. We benchmark RoboKA against strong unimodal and multimodal baselines in both in-domain and out-of-domain setups, finding RoboKA to surpass all baselines in terms of recall and F1-score. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_00156 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RoboKA: KAN Informed Multimodal Learning for RoboCall Surveillance System Choudhury, Nitin Kumar, Nikhil Sinha, Aditya Kumar Anand, Abhijeet Salemi, Hossein Phukan, Orchid Chetia Purohit, Hemant Buduru, Arun Balaji Multimedia Cryptography and Security Wide exploration on robocall surveillance research is hindered due to limited access to public datasets, due to privacy concerns. In this work, we first curate Robo-SAr, a synthetic robocall dataset designed for robocall surveillance research. Robo-SAr comprises of ~200 unwanted and ~1200 legitimate synthetic robocall samples across three realistic adversarial axes: psycholinguistics-manipulated transcripts, emotion-eliciting speech, and cloned voices. We further propose RoboKA, a Kolmogorov-Arnold Network (KAN)-based multimodal fusion framework designed to model structured nonlinear interactions between acoustic and linguistic cues that characterize diverse adversarial robocall strategies. RoboKA first leverages cross-modal contrastive learning to align latent modality representations and feeds the resulting embeddings to a KAN-projection head for final classification. We benchmark RoboKA against strong unimodal and multimodal baselines in both in-domain and out-of-domain setups, finding RoboKA to surpass all baselines in terms of recall and F1-score. |
| title | RoboKA: KAN Informed Multimodal Learning for RoboCall Surveillance System |
| topic | Multimedia Cryptography and Security |
| url | https://arxiv.org/abs/2605.00156 |