A patent application titled "Search With Stateful Chat" appears to describe functionalities similar to Google's Gemini app for smartphones. Additionally, several other patent applications detail experimental processes related to large language models, including text ranking, user embedding for personalization, and prompt-based query generation for retrieval systems. These patents reference research papers on topics like pairwise ranking prompting, user embedding models, and a few-shot dense retrieval method called PROMPTAGATOR. AI
IMPACT These patents and research papers highlight ongoing advancements in LLM capabilities for search, personalization, and query generation, potentially influencing future AI product development.
RANK_REASON The cluster discusses patent applications and associated research papers detailing experimental LLM processes, rather than a direct product release or significant industry event. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — mastodon.social →
- arxiv.org/pdf/2209.11755
- arxiv.org/pdf/2306.17563
- arxiv.org/pdf/2401.04858
- Gemini
- Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting
- Method for Text Ranking with Pairwise Ranking Prompting
- PROMPTAGATOR
- Promptagator - Few-shot Dense Retrieval from 8 Examples
- Search With Stateful Chat
- Systems and methods for prompt-based query generation for diverse retrieval
- US20240289407A1
- User Embedding Model for Personalized Language Prompting
- User Embedding Models for Personalization of Sequence Processing Models
- WO2024064249A1
- WO2025102041A1
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
