ComplexQuestions

ComplexQuestions^[1] consists of 2100 multi-constraint question-answer pairs coming from 3 sources.

This dataset can be downloaded via the link.

Leaderboard

Model / System	Year	Average F1	Reported by
Lan and Jiang [1]	2020	43.3	Yonghui Jia and Wenliang Chen
Reranking	2022	42.9	Yonghui Jia and Wenliang Chen
Luo et al. [2]	2020	42.8	Yonghui Jia and Wenliang Chen
Wang et al.	2022	42.6	Wang et al.
(Bao et al.)’s work	2018	42.33	Bao et. al.
SGM	2022	41.38	Ma L et al.
MulCG	2016	40.94	Ma L et al.
Bao et al. [3]	2016	40.9	Yonghui Jia and Wenliang Chen
SeqM	2020	40.55	Ma L et al.
Ranking	2022	38.4	Yonghui Jia and Wenliang Chen
STAGG	2016	37.69	Bao et. al.
STAGG	2016	36.89	Ma L et al.
UHop	2016	35.30	Ma L et al.

References

[Dataset] Bao, Junwei, Nan Duan, Zhao Yan, Ming Zhou, and Tiejun Zhao. Constraint-based question answering with knowledge graph. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pp. 2503-2514. 2016.

• [1] Lan, Y., Jiang, J.: Query graph generation for answering multi-hop complex questions from knowledge bases. In: Proceedings of ACL. pp. 969–974 (2020)
• [2] Luo,K.,Lin,F.,Luo,X.,Zhu,K.:Knowledge base question answering via encoding of complex query graphs. In: Proceedings of EMNLP. pp. 2185–2194 (2018)
• [3] Bao, J., Duan, N., Yan, Z., Zhou, M., Zhao, T.: Constraint-based question answer- ing with knowledge graph. In: Proceedings of COLING. pp. 2503–2514 (2016)

Note that MulCG, (Bao et al.)’s work and Bao et al. [3] are the same system reported by different papers.

Go back to the README