Brief

A new study evaluated 22 different models, ranging from small encoders to large instruction-tuned LLMs, on their ability to process patent data for tasks like retrieval, classification, and clustering. The research found that fine-tuning effectiveness is highly dependent on the specific task and that gains in one area do not always transfer to others. While larger models generally performed better within their families, cross-family comparisons showed noisy results, with smaller models sometimes outperforming larger ones on specific tasks. The study also highlighted that combining abstract and claim information significantly improved retrieval and classification, though all models struggled with out-of-domain queries. AI

Researchers investigated the effectiveness of synthetic data generated by large language models for low-resource multi-label patent classification. Their findings indicate that while synthetic data can improve classification performance, much of the gain is attributable to increased data volume rather than true synthetic value. The study also revealed that the correlation between fidelity metrics and classification gain varies significantly with data scarcity, and optimal data mixing strategies depend on the generation method. AI