Small specialized llms improve wastewater treatment science and sustainability
Scientific capabilities and deployment sustainability of small-scale LLMs in biological wastewater treatment
Computational Engineering, Finance, and ScienceComputers and Society
Summary
Large language models (LLMs) usually need a lot of computing power and general knowledge, which makes them hard to use effectively in environmental fields like wastewater treatment. The authors created a smaller LLM called BioWater, specialized in this area, that matches or outperforms bigger general LLMs and even experts in some tasks. BioWater helped generate a new scientific idea that experiments later supported, showing its real research potential. Using smaller LLMs locally also saves energy and money compared to big cloud-based models when used extensively in wastewater plants.
What this means in practice
- •For wastewater plant operators: Deploy small specialized LLMs locally to improve scientific insight and operational decisions sustainably in biological wastewater treatment.
- •For environmental ai engineers: Build and fine-tune compact LLMs specialized in environmental domains to reduce computation and energy costs while maintaining scientific capabilities.
Authors
Run-Ze Xu, Chu-Kuan Jiang, Dylan Ming-Han Li, Hong-Xiao Guo, Jia-Shun Cao, Guang-Hao Chen
Abstract
Large language models (LLMs) are emerging as scientific assistants, yet their computational demands and limited domain specialization constrain sustainable deployment in environmental engineering. Here, we investigate whether domain-specialized small-scale LLMs can combine scientific capability with sustainable deployment in biological wastewater treatment. We developed a benchmark evaluating three scientific capabilities of LLMs: retrospective cognition, comprehension fidelity, and prospective extrapolation. BioWater (8 billion parameters, fine-tuned on specialized domain knowledge) achieved higher comprehension-fidelity scores than participating human experts and performance comparable to a 397-billion-parameter general-purpose LLM in retrospective cognition and prospective extrapolation. Human-BioWater collaboration generated a scientific hypothesis that was subsequently supported by laboratory experiments, demonstrating its potential to contribute to prospective scientific research. We further evaluated the economic and environmental implications of LLM deployment across global wastewater treatment plants (WWTPs). Locally deployed small-scale LLMs became more sustainable than cloud-based large-scale LLMs as inference demand increased in intelligent WWTPs. These findings highlight domain-specialized small-scale LLMs as a promising pathway towards scientifically capable, computationally efficient, and sustainably deployable artificial intelligence for wastewater treatment.