FB-Bench: A Fine-Grained Multi-Task Benchmark for Evaluating LLMs' Responsiveness to Human Feedback

TL;DR

FB-Bench is a comprehensive multi-task benchmark designed to evaluate how well large language models respond to nuanced human feedback in realistic, multi-turn Chinese dialogue scenarios, highlighting current strengths and gaps.

Contribution

This paper introduces FB-Bench, a detailed multi-task benchmark for assessing LLM responsiveness to human feedback in complex, real-world interactions, especially in Chinese language contexts.

Findings

Significant performance variation across models and scenarios.

Task type and feedback quality greatly influence responsiveness.

Current models show both strengths and notable limitations.

Abstract

Human feedback is crucial in the interactions between humans and Large Language Models (LLMs). However, existing research primarily focuses on benchmarking LLMs in single-turn dialogues. Even in benchmarks designed for multi-turn dialogues, the user inputs are often independent, neglecting the nuanced and complex nature of human feedback within real-world usage scenarios. To fill this research gap, we introduce FB-Bench, a fine-grained, multi-task benchmark designed to evaluate LLMs' responsiveness to human feedback under real-world usage scenarios in Chinese. Drawing from the two main interaction scenarios, FB-Bench comprises 591 meticulously curated samples, encompassing eight task types, five deficiency types of response, and nine feedback types. We extensively evaluate a broad array of popular LLMs, revealing significant variations in their performance across different interaction scenarios. Further analysis indicates that task, human feedback, and deficiencies of previous responses can also significantly impact LLMs' responsiveness. Our findings underscore both the strengths and limitations of current models, providing valuable insights and directions for future research. Code and datasets are available at https://github.com/PKU-Baichuan-MLSystemLab/FB-Bench.