Searcharxiv⌕ Search

arXiv subjects

Yasaman Mohammadpour

Publications and source records attributed to Yasaman Mohammadpour.

3 recordsLinked to original sources

HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition

Self-supervised learning for wearable sensing is usually confined to a single dataset or device family, leaving open whether representations transfer across heterogeneous sensor configurations and whether gains come from pretraining rather than architecture or richer sensor input. We present HeteroWear, a sensor-aware self-supervised framework pretrained on an eight-dataset wearable corpus (173 subjects, 358,261 windows) spanning inertial and physiological signals across body locations, devices, and sampling rates. HeteroWear separates within-sensor temporal modeling from cross-sensor relational modeling using sensor-context embeddings, temporal and cross-sensor Transformers, and latent cross-attention fusion. Training combines structured sensor dropout, cross-configuration consistency, and per-axis masked sensor modeling; missing sensors are omitted rather than reconstructed. Under leave-one-dataset-out pretraining and subject-disjoint evaluation on PAMAP2, WISDM, HHAR, HARTH, and UCI-HAR, HeteroWear improves mean macro-F1 from 0.688 to 0.724 over the identical architecture trained from scratch, with the largest gain on HHAR (0.823 versus 0.741), and achieves the highest macro-F1 on four of five benchmarks relative to single-accelerometer supervised baselines. Held-out-device evaluation yields a mean +0.066 macro-F1 improvement. Under 50% test-time channel removal, HeteroWear retains 73.8% of full-channel performance versus 64.1% for the same architecture trained from scratch without sensor dropout. Ablations show that structured sensor dropout primarily drives robustness, while cross-configuration consistency and sensor-context encoding contribute to transfer.

cs.LG↗

Sentiment and Social Signals in the Climate Crisis: A Survey on Analyzing Social Media Responses to Extreme Weather Events

Extreme weather events driven by climate change, such as wildfires, floods, and heatwaves, prompt significant public reactions on social media platforms. Analyzing the sentiment expressed in these online discussions can offer valuable insights into public perception, inform policy decisions, and enhance emergency responses. Although sentiment analysis has been widely studied in various fields, its specific application to climate-induced events, particularly in real-time, high-impact situations like the 2025 Los Angeles forest fires, remains underexplored. In this survey, we thoroughly examine the methods, datasets, challenges, and ethical considerations related to sentiment analysis of social media content concerning weather and climate change events. We present a detailed taxonomy of approaches, ranging from lexicon-based and machine learning models to the latest strategies driven by large language models (LLMs). Additionally, we discuss data collection and annotation techniques, including weak supervision and real-time event tracking. Finally, we highlight several open problems, such as misinformation detection, multimodal sentiment extraction, and model alignment with human values. Our goal is to guide researchers and practitioners in effectively understanding sentiment during the climate crisis era.

cs.SI↗

Explainable Human-in-the-Loop Segmentation via Critic Feedback Signals

Segmentation models achieve high accuracy on benchmarks but often fail in real-world domains by relying on spurious correlations instead of true object boundaries. We propose a human-in-the-loop interactive framework that enables interventional learning through targeted human corrections of segmentation outputs. Our approach treats human corrections as interventional signals that show when reliance on superficial features (e.g., color or texture) is inappropriate. The system learns from these interventions by propagating correction-informed edits across visually similar images, effectively steering the model toward robust, semantically meaningful features rather than dataset-specific artifacts. Unlike traditional annotation approaches that simply provide more training data, our method explicitly identifies when and why the model fails and then systematically corrects these failure modes across the entire dataset. Through iterative human feedback, the system develops increasingly robust representations that generalize better to novel domains and resist artifactual correlations. We demonstrate that our framework improves segmentation accuracy by up to 9 mIoU points (12-15\% relative improvement) on challenging cubemap data and yields 3-4$\times$ reductions in annotation effort compared to standard retraining, while maintaining competitive performance on benchmark datasets. This work provides a practical framework for researchers and practitioners seeking to build segmentation systems that are accurate, robust to dataset biases, data-efficient, and adaptable to real-world domains such as urban climate monitoring and autonomous driving.

cs.CV↗