Neuro-symbolic AI

I believe a neuro-symbolic approach is the only way to realize truthworthy reasoning, thus achieving true intelligence. It is probably the only hope to bridge the seemingly insurmountable gap between “correlation” and “causality”.

Here are some useful cases of the neural symbolic approach. And I think the role of a neural-symbolic engine will become much more important in the near future.

Synthesizing high-quality, verifiably correct reasoning steps on solving math problems. (Li et al., 2024).
Improving the autoformalization, a fundamental math capability of a neural model, through a hybrid neural-symbolic solution. (Li et al., 2024).

References

2024

NeurIPS

Neuro-Symbolic Data Generation for Math Reasoning

Zenan Li, and 7 more authors

In Advances in Neural Information Processing Systems, NeurIPS, 2024

Bib HTML

@inproceedings{NEURIPS2024_NSMR,
  title = {Neuro-Symbolic Data Generation for Math Reasoning},
  author = {Li, Zenan and Zhou, Zhi and Yao, Yuan and Zhang, Xian and Li, Yu-Feng and Cao, Chun and Yang, Fan and Ma, Xiaoxing},
  year = {2024},
  booktitle = {Advances in Neural Information Processing Systems, {NeurIPS}},
}

NeurIPS

Autoformalize Mathematical Statements by Symbolic Equivalence and Semantic Consistency

Zenan Li, and 6 more authors

In Advances in Neural Information Processing Systems, NeurIPS, 2024

Bib HTML

@inproceedings{NEURIPS2024_AutoFormalization,
  title = {Autoformalize Mathematical Statements by Symbolic Equivalence and Semantic Consistency},
  author = {Li, Zenan and Wu, Yifan and Li, Zhaoyu and Wei, Xinming and Zhang, Xian and Yang, Fan and Ma, Xiaoxing},
  year = {2024},
  booktitle = {Advances in Neural Information Processing Systems, {NeurIPS}},
}