New AI method improves patent claim generation with topology and content decoding

Researchers have developed a new method called SPG (Structure-aware Patent Generation) to improve the autoregressive generation of patent claims. This method addresses the limitation of flat token sequences in standard decoders by jointly decoding topology and content, allowing for hierarchical constraints and monotonic scope narrowing. SPG incorporates a pointer head to select parent claims and a depth-adaptive scope regularizer, enhancing antecedent consistency and recovering a significant portion of gold parent links. AI

IMPACT This research could lead to more accurate and structured AI generation of complex legal documents like patents.

RANK_REASON The cluster contains a research paper detailing a novel method for AI-driven patent claim generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• HUPD-DCG
• Influence Flower
• LLaMA-3-8B-Instruct
• ScienceCast

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →