Advancements in Automated Scholarly Content Generation and Analysis

The field of automated scholarly content generation and analysis is witnessing significant advancements, driven by innovations in artificial intelligence, natural language processing, and machine learning. Researchers are developing novel systems and frameworks that can generate high-quality academic surveys, system architecture diagrams, and scientific posters, as well as analyze and understand the structure and content of these artifacts. Furthermore, there is a growing focus on addressing the disparities and biases in data science and artificial intelligence, particularly in terms of gender and diversity. Noteworthy papers in this area include ARISE, which introduces an agentic rubric-guided iterative survey engine for automated generation and refinement of academic survey papers, and RhinoInsight, which presents a deep research framework that enhances robustness, traceability, and overall quality of model behavior and context. Other notable works include Paper2SysArch, which establishes a foundational benchmark for automated scientific visualization, and DR Tulu, which develops a reinforcement learning approach for deep research with evolving rubrics. These advancements have the potential to revolutionize the way researchers work and collaborate, and to promote greater diversity, equity, and inclusion in the field.

Sources

ARISE: Agentic Rubric-Guided Iterative Survey Engine for Automated Scholarly Paper Generation

Paper2SysArch: Structure-Constrained System Architecture Generation from Scientific Papers

SciPostLayoutTree: A Dataset for Structural Analysis of Scientific Posters

Bridging the Divide: Gender, Diversity, and Inclusion Gaps in Data Science and Artificial Intelligence Across Academia and Industry in the majority and minority worlds

RhinoInsight: Improving Deep Research through Control Mechanisms for Model Behavior and Context

Data Flows and Colonial Regimes in Africa: A Critical Analysis of the Colonial Futurities Embedded in AI Ecosystems

PRInTS: Reward Modeling for Long-Horizon Information Seeking

UISearch: Graph-Based Embeddings for Multimodal Enterprise UI Screenshots Retrieval

DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

CodeR3: A GenAI-Powered Workflow Repair and Revival Ecosystem

Invisible in Search? Auditing Aesthetic Bias in the Visual Representation of Holocaust Victims on Google

Built with on top of