JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning

TL;DR

JsonTuning introduces a structured approach using JSON to improve instruction tuning of large language models, enhancing their generalization, robustness, and controllability over traditional text-based methods.

Contribution

The paper proposes JsonTuning, a novel structure-to-structure instruction tuning method that explicitly encodes task information to outperform existing text-to-text approaches.

Findings

JsonTuning outperforms TextTuning in various benchmarks.

JsonTuning improves model robustness and controllability.

Extensive analysis confirms the effectiveness of JSON-based task representation.

Abstract

Instruction tuning is vital for enhancing the performance of large language models (LLMs), but existing text-to-text methods, referred to as TextTuning, struggle with issues such as generalization, robustness, and controllability due to their lack of explicit task structures. We introduce JsonTuning, a structure-to-structure approach that uses JSON structures to represent tasks. This method improves generalization by clarifying task elements and their relations, boosts robustness by minimizing ambiguity, and enhances controllability by allowing precise control over outputs. We conduct an extensive comparative analysis between JsonTuning and TextTuning using various language models and benchmarks. Our findings reveal that JsonTuning consistently surpasses TextTuning in terms of performance, robustness, and controllability across different scenarios. By overcoming the limitations of TextTuning, JsonTuning demonstrates significant potential for developing more effective and reliable LLMs capable of handling diverse scenarios.