PulseAugur / Brief
EN
LIVE 11:04:15

Brief

last 24h
[2/2] 224 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

  1. ScholaWrite: A Dataset of End-to-End Scholarly Writing Process

    Researchers have introduced ScholaWrite, a novel dataset designed to capture the complete scholarly writing process. This dataset was collected using a Chrome extension that recorded keystrokes within Overleaf, documenting the multi-month journey from initial drafts to final manuscripts for five computer science preprints. The data includes over 62,000 text changes and provides insights into the cognitive demands and task-switching involved in academic writing, highlighting current limitations of LLMs in assisting this process. AI

    IMPACT Provides data to develop more effective AI writing assistants that understand the cognitive process of authors.

  2. PaperMentor: A Human-Centered Multi-Agent Writing Tutor for AI Research Papers on Overleaf

    Researchers have developed PaperMentor, a novel writing assistant designed to help early-career scholars improve their AI research papers. This human-centered system integrates expert advice with specialized agents to provide actionable feedback directly within Overleaf, distinguishing itself from basic grammar checkers. In a user study, PaperMentor's comments were rated highly actionable and valid, significantly outperforming a GPT-5.2 baseline. AI

    IMPACT Provides a specialized AI tool to improve the quality and efficiency of academic writing for AI researchers.