SciGLM: Training Scientific Language Models with Self-Reflective Instruction Annotation and Tuning

Anonymous

Abstract: Large Language Models (LLMs) have shown promise in assisting scientific discovery. However, such applications are currently limited by LLMs' deficiencies in understanding intricate scientific concepts, deriving symbolic equations, and solving advanced numerical calculations. To bridge these gaps, we introduce SciGLM, a suite of scientific language models able to conduct college-level scientific reasoning. Central to our approach is a novel self-reflective instruction annotation framework to address the data scarcity challenge in the science domain. This framework leverages existing LLMs to generate step-by-step reasoning for unlabelled scientific questions, followed by a process of self-reflective critic-and-revise. Applying this framework, we curated SciInstruct, a diverse and high-quality dataset encompassing physics, chemistry, math, and formal proofs. We fine-tuned the ChatGLM family of language models with SciInstruct, enhancing their capabilities in scientific and mathematical reasoning. Remarkably, the SciGLM consistently improves both the base model (ChatGLM3-6B-Base) and larger-scale models (32B), without sacrificing the language understanding capabilities of the base model. This makes SciGLM a suitable foundational model to facilitate diverse scientific discovery tasks. For the benefit of the wider research community, we release SciInstruct, and SciGLM, alongside a self-reflective framework and fine-tuning code.

Paper Type: long

Research Area: NLP Applications

Contribution Types: Publicly available software and/or pre-trained models, Data resources

Languages Studied: scientific language model, self-reflective instruction, scientific reasoning