Effective Distillation of Table-based Reasoning Ability from LLMs
This addresses the challenge of deploying LLMs in resource-constrained settings by enabling efficient table reasoning in smaller models, though it is incremental as it builds on existing distillation and table reasoning research.
The paper tackles the problem of distilling table-based reasoning ability from large language models (LLMs) into smaller models for scientific table-to-text generation, achieving a 220 million parameter model that outperforms both traditional fine-tuned baselines and specific LLMs on a dataset.
Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of natural language processing tasks. However, their enormous parameter size and extremely high requirements for compute power pose challenges for their practical deployment. Recent research has revealed that specific capabilities of LLMs, such as numerical reasoning, can be transferred to smaller models through distillation. Some studies explore the potential of leveraging LLMs to perform table-based reasoning. However, there has been no prior work focusing on table reasoning skills in smaller models specifically tailored for scientific table-to-text generation tasks. In this paper, we propose a novel table-based reasoning distillation approach, with the aim of distilling LLMs into tailored smaller models. Our experimental results have shown that a 220 million parameter model (Flan-T5-base) fine-tuned using distilled data, not only achieves a significant improvement compared to traditionally fine-tuned baselines, but also surpasses specific LLMs on a scientific table-to-text generation dataset. Our code is available at https://github.com/Bernard-Yang/DistillTableCoT.