Researchers have developed a method to teach language models to improve Boolean patent searches by learning from expert feedback. By analyzing 38,397 Japan Patent Office search reports, they created strategies that generate valid, high-recall queries. A novel line-level learning approach was particularly effective, significantly reducing the size of document sets containing relevant citations, making screening more efficient. AI
IMPACT Enhances efficiency in specialized information retrieval tasks like patent searching.
RANK_REASON Academic paper detailing a new method for language models. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- CatalyzeX
- Connected Papers
- DagsHub
- Gotit.pub
- Hugging Face
- Japan Patent Office
- Litmaps
- ScienceCast
- Scite
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →