PAPERWEAVER is an enriched paper alerts system that provides contextualized text descriptions of recommended papers based on user-collected papers. It uses Large Language Models (LLMs) to infer users' research interests from their collected papers, extract context-specific aspects of papers, and compare recommended and collected papers on these aspects. A user study with 15 participants showed that users using PAPERWEAVER were able to better understand the relevance of recommended papers and triage them more confidently compared to a baseline that presented related work sections from recommended papers. PAPERWEAVER generates contextualized aspect-based summaries and paper-paper descriptions to help users understand how recommended papers relate to their research context. It also anchors unfamiliar papers with familiar collected papers to reduce cognitive load. The system is built on an existing document recommender system and leverages LLMs for text generation. PAPERWEAVER provides three types of descriptions: contextualized aspect-based summaries, paper-paper descriptions based on citances, and paper-paper descriptions via generated pseudo-citances. The system allows users to explore multiple descriptions for recommended papers and helps them understand the nuanced relevance of the papers to their research context. The paper presents the design and evaluation of PAPERWEAVER, which aims to improve the understanding and triage of recommended papers by providing contextualized descriptions. The study shows that PAPERWEAVER helps users better understand the relevance of recommended papers and their relationships to collected papers. The system is designed to support scholars in managing an ever-increasing accumulation of knowledge by providing contextualized descriptions of recommended papers.PAPERWEAVER is an enriched paper alerts system that provides contextualized text descriptions of recommended papers based on user-collected papers. It uses Large Language Models (LLMs) to infer users' research interests from their collected papers, extract context-specific aspects of papers, and compare recommended and collected papers on these aspects. A user study with 15 participants showed that users using PAPERWEAVER were able to better understand the relevance of recommended papers and triage them more confidently compared to a baseline that presented related work sections from recommended papers. PAPERWEAVER generates contextualized aspect-based summaries and paper-paper descriptions to help users understand how recommended papers relate to their research context. It also anchors unfamiliar papers with familiar collected papers to reduce cognitive load. The system is built on an existing document recommender system and leverages LLMs for text generation. PAPERWEAVER provides three types of descriptions: contextualized aspect-based summaries, paper-paper descriptions based on citances, and paper-paper descriptions via generated pseudo-citances. The system allows users to explore multiple descriptions for recommended papers and helps them understand the nuanced relevance of the papers to their research context. The paper presents the design and evaluation of PAPERWEAVER, which aims to improve the understanding and triage of recommended papers by providing contextualized descriptions. The study shows that PAPERWEAVER helps users better understand the relevance of recommended papers and their relationships to collected papers. The system is designed to support scholars in managing an ever-increasing accumulation of knowledge by providing contextualized descriptions of recommended papers.