How to "Extract aligned spans from a Question and an Answer"?

[tin-xai]

Hi, I found LangExtract is really relevant to my work HoT (https://highlightedchainofthought.github.io/) that tries to extract semantically similar phrases between question and answer. And my method improve the accuracy over multiple prompting methods as well.

I did some modification to make LangExtract work with HoT, and it works pretty well:

If you find this useful, please cite our work (HoT) and LangExtract: @article{nguyen2025hot, title={HoT: Highlighted Chain of Thought for Referencing Supporting Facts from Inputs}, author={Nguyen, Tin and Bolton, Logan and Taesiri, Mohammad Reza and Nguyen, Anh Totti}, journal={arXiv preprint arXiv:2503.02003}, year={2025} }

LangExtract. (2024). Google Research. Retrieved from https://github.com/google/langextract

[tin-xai]

Link to this work: https://github.com/tin-xai/HoT_with_LangExtract

[aksg87]

Thanks for sharing your library @tin-xai - Looks like very interesting work!