Expert-guided and LLM-augmented feature selection improved generalizability of data-driven models for nitrous oxide emission prediction at a full-scale wastewater plant, maintaining temporal dynamics under out-of-distribution high-flow conditions where attention-based deep learning failed.
By the publication date of 2026-07-12, researchers tested a knowledge-driven feature selection framework for data-driven wastewater modeling, comparing classic attention-based deep learning against expert-guided and LLM-augmented selection. In the reported case study of N2O emissions at a full-scale plant, expert-guided selection achieved mean R2 0.723 and MAE 0.033, slightly above the best attention model at R2 0.712, while LLM-augmented reached R2 0.596 and MAE 0.041.
- Impact 30%
- 49
- Evidence 25%
- 95
- Scale 20%
- 35
- Confidence 15%
- 87
- Recency 10%
- 83
Updated Jul 13, 2026 · TRV-2026-0120
ace