Research map / Label-efficient and robust learning / Recognition and long-tail
HumBugDB: A Large-scale Acoustic Mosquito Dataset
Ivan Kiskin, Marianne Sinka, Adam D. Cobb, Waqas Rafique, Lawrence Wang, Davide Zilli, Benjamin Gutteridge, Rinita Dam, Theodoros Marinos, Yunpeng Li, Dickson Msaky, Emmanuel Kaindoa and 4 more
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Feat. A instead of Feat. B | MozzBNNv2 · MED Test A | ROC AUC: 96.4 -> 98.1 | Improved |
| use Feat. A instead of Feat. B | BNN-ResNet50 · MED Test A | ROC AUC: 93.0 -> 98.3 | Improved |
| use Feat. A instead of Feat. B | BNN-ResNet18 · MED Test A | ROC AUC: 92.5 -> 98.1 | Improved |
| use Feat. A instead of Feat. B | BNN-VGGish · MED Test A | ROC AUC: 97.3 -> 98.5 | Improved |
| use Feat. A instead of Feat. B | BNN-ResNet50 · MED Test B | ROC AUC: 76.1 -> 74.8 | Hurt |
| use Feat. A instead of Feat. B | MozzBNNv2 · MED Test B | ROC AUC: 58.4 -> 71.1 | Improved |
| use Feat. A instead of Feat. B | BNN-VGGish · MED Test B | ROC AUC: 57.4 -> 74.1 | Improved |
| evaluate on Test B (low SNR, unseen domain) instead of Test A | BNN-ResNet50 Feat. A · MED | TPR: 76.9 -> 31.0 | Hurt |
| train on half of Test B and test on the other half | BNN model · MED Test B | ROC AUC: 0.770 -> 0.915 | Improved |
| use Feat. B instead of Feat. A | MozzBNNv2 · MSC total | ROC AUC: 91.4 -> 92.7 | Improved |
| use Feat. B instead of Feat. A | MozzBNNv2 · MSC total | PR AUC: 66.9 -> 71.6 | Improved |
| use Feat. B instead of Feat. A | BNN-VGGish · MSC total | ROC AUC: 92.1 -> 91.4 | Hurt |
Closest papers by meaning
- The iNaturalist Sounds Dataset (NeurIPS 2024 · similarity 0.84)
- BIOSCAN-5M: A Multimodal Dataset for Insect Biodiversity (NeurIPS 2024 · similarity 0.83)
- Open-Insect: Benchmarking Open-Set Recognition of Novel Species in Biodiversity Monitoring (NeurIPS 2025 · similarity 0.82)
- Insect-Foundation: A Foundation Model and Large-scale 1M Dataset for Visual Insect Understanding (CVPR 2024 · similarity 0.81)
- Noisy Ostracods: A Fine-Grained, Imbalanced Real-World Dataset for Benchmarking Robust Machine Learning and Label Correction Methods (NeurIPS 2024 · similarity 0.81)
- NatureLM-audio: an Audio-Language Foundation Model for Bioacoustics (ICLR 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
