Systematic Task Exploration with LLMs: A Study in Citation Text Generation

TL;DR

This paper introduces a systematic framework for exploring creative natural language generation tasks using large language models, focusing on citation text generation, and highlights the importance of input and instruction design in model performance and evaluation.

Contribution

It proposes a three-component research framework for task exploration with LLMs and applies it to citation text generation, providing new insights and evaluation methods.

Findings

Systematic input manipulation affects LLM performance.

Evaluation metrics have complex relationships in citation generation.

Human evaluation offers qualitative insights into task challenges.

Abstract

Large language models (LLMs) bring unprecedented flexibility in defining and executing complex, creative natural language generation (NLG) tasks. Yet, this flexibility brings new challenges, as it introduces new degrees of freedom in formulating the task inputs and instructions and in evaluating model performance. To facilitate the exploration of creative NLG tasks, we propose a three-component research framework that consists of systematic input manipulation, reference data, and output measurement. We use this framework to explore citation text generation -- a popular scholarly NLP task that lacks consensus on the task definition and evaluation metric and has not yet been tackled within the LLM paradigm. Our results highlight the importance of systematically investigating both task instruction and input configuration when prompting LLMs, and reveal non-trivial relationships between different evaluation metrics used for citation text generation. Additional human generation and human evaluation experiments provide new qualitative insights into the task to guide future research in citation text generation. We make our code and data publicly available.