Research

Paper

AI LLM March 09, 2026

Gradually Excavating External Knowledge for Implicit Complex Question Answering

Authors

Chang Liu, Xiaoguang Li, Lifeng Shang, Xin Jiang, Qun Liu, Edmund Y. Lam, Ngai Wong

Abstract

Recently, large language models (LLMs) have gained much attention for the emergence of human-comparable capabilities and huge potential. However, for open-domain implicit question-answering problems, LLMs may not be the ultimate solution due to the reasons of: 1) uncovered or out-of-date domain knowledge, 2) one-shot generation and hence restricted comprehensiveness. To this end, this work proposes a gradual knowledge excavation framework for open-domain complex question answering, where LLMs iteratively and actively acquire external information, and then reason based on acquired historical knowledge. Specifically, during each step of the solving process, the model selects an action to execute, such as querying external knowledge or performing a single logical reasoning step, to gradually progress toward a final answer. Our method can effectively leverage plug-and-play external knowledge and dynamically adjust the strategy for solving complex questions. Evaluated on the StrategyQA dataset, our method achieves 78.17% accuracy with less than 6% parameters of its competitors, setting new SOTA for ~10B-scale LLMs.

Metadata

arXiv ID: 2603.08148
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-09
Fetched: 2026-03-10 05:43

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.08148v1</id>\n    <title>Gradually Excavating External Knowledge for Implicit Complex Question Answering</title>\n    <updated>2026-03-09T09:28:42Z</updated>\n    <link href='https://arxiv.org/abs/2603.08148v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.08148v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Recently, large language models (LLMs) have gained much attention for the emergence of human-comparable capabilities and huge potential. However, for open-domain implicit question-answering problems, LLMs may not be the ultimate solution due to the reasons of: 1) uncovered or out-of-date domain knowledge, 2) one-shot generation and hence restricted comprehensiveness. To this end, this work proposes a gradual knowledge excavation framework for open-domain complex question answering, where LLMs iteratively and actively acquire external information, and then reason based on acquired historical knowledge. Specifically, during each step of the solving process, the model selects an action to execute, such as querying external knowledge or performing a single logical reasoning step, to gradually progress toward a final answer. Our method can effectively leverage plug-and-play external knowledge and dynamically adjust the strategy for solving complex questions. Evaluated on the StrategyQA dataset, our method achieves 78.17% accuracy with less than 6% parameters of its competitors, setting new SOTA for ~10B-scale LLMs.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-09T09:28:42Z</published>\n    <arxiv:comment>13 pages, 3 figures, EMNLP findings 2023</arxiv:comment>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Chang Liu</name>\n    </author>\n    <author>\n      <name>Xiaoguang Li</name>\n    </author>\n    <author>\n      <name>Lifeng Shang</name>\n    </author>\n    <author>\n      <name>Xin Jiang</name>\n    </author>\n    <author>\n      <name>Qun Liu</name>\n    </author>\n    <author>\n      <name>Edmund Y. Lam</name>\n    </author>\n    <author>\n      <name>Ngai Wong</name>\n    </author>\n  </entry>"
}