Automating mundane tasks keeps your attention focused on the work that matters.

We propose SAmple-centric In-context Learning (SAIL), which introduces a fine-grained entity-level textual similarity to facilitate in-depth text analysis by LLMs and incorporates layout similarity to ...

This is the repo for the RuleR project, which proposes a data augmentation method incorporating multiple constraints into the original data samples according to predefined rules, without any human/LLM ...