BioDCASE keynote: optimally allocating scarce expert time for bioacoustic data labeling
sarameghanbeery · x · 2026-10-28
A keynote at the BioDCASE workshop on computational bioacoustics, covering recent work led by Rupa Vendhan and Timm Haucke on how human verifications affect bioacoustic data labeling quality. The core question: given limited expert time, how to optimally allocate verification effort toward downstream goals like species occupancy, abundance, or richness estimation. Talk video is available.
Related event: Sara Beery Team Proposes Transductive Active Labeling for Bioacoustic Data(3 posts)→
More from Research
- AI designs 35,000 antibodies in weeks, opening new paths for food allergy treatment — SnackSafely · 2026-10-28
- Revisiting Boolean function learnability on deep neural networks — luislamb · 2026-10-28
- CATCH: ~200-example adversarial training makes LLMs honest, beating 50× larger SFT — Mila_Quebec · 2026-10-28
- Avalon-M1 trains LLM agent organisations via collective-reward RL, lifting avg score 33.5% to 45.7% — RobertTLange · 2026-10-28
- Google open-sources EmbeddingGemma 2: a 744M multimodal embedding model beating models 12x larger — dair_ai · 2026-10-28
- Dev trains a 21B MoE (3.36B active) from scratch on 70B tokens and logs the emergence — Prestigious-Taste-63 · 2026-10-28