Saved in:
Bibliographic Details
Main Authors: Hou, Yutao, Luo, Yajing, Ruan, Zhiwen, Wang, Hongru, Ge, Weifeng, Chen, Yun, Chen, Guanhua
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.10163
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911404718555136
author Hou, Yutao
Luo, Yajing
Ruan, Zhiwen
Wang, Hongru
Ge, Weifeng
Chen, Yun
Chen, Guanhua
author_facet Hou, Yutao
Luo, Yajing
Ruan, Zhiwen
Wang, Hongru
Ge, Weifeng
Chen, Yun
Chen, Guanhua
contents Large language models (LLMs) demonstrate remarkable performance across various tasks, prompting researchers to develop diverse evaluation benchmarks. However, most benchmarks typically measure the ability of LLMs to respond to individual questions, neglecting the complex interactions in real-world applications. We introduce Compound Question Synthesis (CQ-Syn) to build Compound-QA, a benchmark targeting questions composed of multiple interrelated sub-questions. This benchmark is derived from existing QA datasets, annotated with proprietary LLMs, and verified by humans for accuracy. It encompasses five categories: Factual-Statement, Cause-and-Effect, Hypothetical-Analysis, Comparison-and-Selection, and Evaluation-and-Suggestion. It evaluates the LLM capability in terms of three dimensions, including understanding, reasoning, and knowledge. Evaluating nine open-source LLMs on Compound-QA reveals that their performance on compound questions is notably lower than on non-compound questions. We further explore strategies to enhance LLMs' handling of compound questions, and our results show that these methods substantially improve models' comprehension and reasoning abilities.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10163
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Compound-QA: A Benchmark for Evaluating LLMs on Compound Questions
Hou, Yutao
Luo, Yajing
Ruan, Zhiwen
Wang, Hongru
Ge, Weifeng
Chen, Yun
Chen, Guanhua
Computation and Language
Large language models (LLMs) demonstrate remarkable performance across various tasks, prompting researchers to develop diverse evaluation benchmarks. However, most benchmarks typically measure the ability of LLMs to respond to individual questions, neglecting the complex interactions in real-world applications. We introduce Compound Question Synthesis (CQ-Syn) to build Compound-QA, a benchmark targeting questions composed of multiple interrelated sub-questions. This benchmark is derived from existing QA datasets, annotated with proprietary LLMs, and verified by humans for accuracy. It encompasses five categories: Factual-Statement, Cause-and-Effect, Hypothetical-Analysis, Comparison-and-Selection, and Evaluation-and-Suggestion. It evaluates the LLM capability in terms of three dimensions, including understanding, reasoning, and knowledge. Evaluating nine open-source LLMs on Compound-QA reveals that their performance on compound questions is notably lower than on non-compound questions. We further explore strategies to enhance LLMs' handling of compound questions, and our results show that these methods substantially improve models' comprehension and reasoning abilities.
title Compound-QA: A Benchmark for Evaluating LLMs on Compound Questions
topic Computation and Language
url https://arxiv.org/abs/2411.10163