Self-Correction Distillation for Structured Data Question Answering
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (19) · 16566-16574
DOI: 10.1609/aaai.v40i19.38697 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yushan Zhu und 10 weiteren Autor:innen.
Abstract (Originalsprache)
Structured data question answering (QA), including table QA, Knowledge Graph (KG) QA, and temporal KG QA, is a pivotal research area. Advances in large language models (LLMs) have driven significant progress in unified structural QA frameworks like TrustUQA. However, these frameworks face challenges when applied to small-scale LLMs since small-scale LLMs are prone to errors in generating structured queries. To improve the structured data QA ability of small-scale LLMs, we propose a self-correction distillation (SCD) method. In SCD, an error prompt mechanism (EPM) is designed to detect errors and provide customized error messages during inference, and a two-stage distillation strategy is designed to transfer large-scale LLMs' query-generation and error-correction capabilities to small-scale LLM. Experiments across 5 benchmarks with 3 structured data types demonstrate that our SCD achieves the best performance and superior generalization on small-scale LLM (8B) compared to other distillation methods, and closely approaches the performance of GPT4 on some datasets. Furthermore, large-scale LLMs equipped with EPM surpass the state-of-the-art results on most datasets.
Zitieren
APA
Zhu, Y., Zhang, W., Jin, L., Sun, M., Zhong, L., Liu, Z., Li, J., Liang, L., Long, C., Deng, C., & Feng, J. (2026). Self-Correction Distillation for Structured Data Question Answering. Proceedings of the AAAI Conference on Artificial Intelligence, 40(19), 16566-16574. https://doi.org/10.1609/aaai.v40i19.38697
Vancouver
Zhu Y, Zhang W, Jin L, Sun M, Zhong L, Liu Z, et al. Self-Correction Distillation for Structured Data Question Answering. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(19):16566-16574. doi:10.1609/aaai.v40i19.38697
BibTeX
@article{Zhu2026,
author = {Zhu, Yushan and Zhang, Wen and Jin, Long and Sun, Mengshu and Zhong, Ling and Liu, Zhiqiang and Li, Juan and Liang, Lei and Long, Chong and Deng, Chao and Feng, Junlan},
title = {Self-Correction Distillation for Structured Data Question Answering},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {19},
pages = {16566--16574},
doi = {10.1609/aaai.v40i19.38697}
}
EUVIMED